General Electric Company v Ausrail Technologies Pty Ltd - [2015] APO 67

IP AUSTRALIA
AUSTRALIAN PATENT OFFICE
General Electric Company v Ausrail Technologies Pty Ltd [2015] APO 67

Patent Application: 2008201906

Title:Method for improving timekeeping and saving energy on long-haul train

Patent Applicant:  Ausrail Technologies Pty Ltd
Opponent:  General Electric Company
Delegate:  Matthew Lee
Decision Date:  19 October 2015
Hearing Date:  5 August 2015, in Canberra
Catchwords:  PATENTS – section 59 – opposition to grant of a patent on the grounds of novelty, inventive step, clarity, fair basis and manner of manufacture – invention directed towards optimal control of a train for energy efficiency – person skilled in the art – some claims lack novelty – non-patent literature not shown to be ascertainable by skilled persons – claims involve an inventive step – claims are clear – invention fairly based – some claims not a manner of manufacture - opposition successful – applicant afforded opportunity to amend – costs awarded against applicant

Representation: Patent applicant: Katrina Crooks and Russell Davies of Shelston IP, Sydney.

Opponent:Ian Horak of Counsel, instructed by Alyssa Grabb and David Longmuir of Philips Ormonde & Fitzpatrick, Melbourne.

IP AUSTRALIA
AUSTRALIAN PATENT OFFICE
Patent Application: 2008201906

Title:Method for improving timekeeping and saving energy on long-haul train

Patent Applicant: Ausrail Technologies Pty Ltd
Date of Decision: 19 October 2015
DECISION
The opposition succeeds on the ground that claims 29-34, 40 and 44 lack novelty in accordance with section 18(1)(b)(i), and claims 1-28 and 42-43 are not a manner of manufacture in accordance with section 18(1)(a). The applicant is allowed 2 months from the date of this decision to file suitable amendments. Costs are awarded against the applicant according to Schedule 8.
REASONS FOR DECISION
Background

Ausrail Technologies Pty Ltd is the applicant for patent application 2008201906 which was filed as a divisional application to 2003229097, claiming a priority date of 20 May 2002. The application was advertised as accepted on 3 December 2009. A notice of opposition to the grant of the patent was filed by General Electric Company on 3 March 2010. A statement of grounds and particulars was filed on 3 June 2010, and amendments to the statement of grounds and particulars were filed on 16 January 2014 and subsequently allowed.
Grounds of Opposition
The grounds of opposition relied upon by the opponent are those available under subsections 59(b) and 59(c) of the Patents Act 1990 as in force immediately before 15 April 2013 (i.e. prior to the amendments made by the Intellectual Property Laws Amendment (Raising the Bar) Act 2012), that is, firstly, that the invention as claimed is not a patentable invention as it does not comply with subsections 18(1)(a), 18(1)(b)(i) and 18(1)(b)(ii) (the claimed invention is not a manner of manufacture, is not novel and does not involve an inventive step), and, secondly, that the specification filed in respect of the complete application does not comply with subsection 40(3) (the claims lack clarity and fair basis).
Any further references to the Patents Act in the following paragraphs will similarly refer to the legislation as in force immediately before 15 April 2013.
Onus of Proof
The onus of proof in opposition proceedings lies with the opponent, who must establish that it is clear that a valid patent cannot be granted (F.Hoffman-La Roche AG v New England Biolabs Inc [2000] FCA 283 at [29], [67]; (2001) 50 IPR 305 at 311 [29], 319 [67]; Commissioner of Patents v Sherman [2008] FCAFC 182 at [18], [22]; (2009) 79 IPR 426 at 430 [18], 432 [22]).
The Specification
The specification relates to a method of operating trains on a rail network, in particular long-haul rail networks, in such a way as to reduce the energy costs of the journey. It is understandably advantageous to reduce energy consumption for train journeys in order to minimise costs, but the specification notes that any such journey would also need to satisfy constraints such as arriving on or close to the desired time in order to maintain efficiency. The specification then identifies other factors that may also constrain the journey such as speed limits which need to be considered when determining the optimal journey.
The specification goes on to state that the present invention provides an improved method of operating trains to provide an energy-efficient journey by monitoring the progress of the train on a long-haul network, calculating efficient control profiles for the train, and displaying driving advice to the train crew. Page 5 lines 10-18 further describes the invention:
“The system performs four main tasks:
i.State estimation: monitors the progress of a journey to determine the current location and speed of the train;
ii.Train parameter estimation: estimates some parameters of a train performance model;
iii.Journey optimisation: calculates or selects an energy-efficient driving strategy that will get the train to the next key location as close as possible to the desired time; and
iv.Advice generation: generates and provides driving advice for the driver.”
The journey optimisation calculated in step (iii) is described at page 6 lines 25-29 of the specification:
“The optimal journey profile between a given journey state and a target journey state is found by solving a set of differential equations for the motion of the train and an additional differential equation that determines the optimal control. The optimal journey profile specifies the time, speed and control at each location of the track and between the current train location and the next target.”
The specification continues by describing a series of equations, and in particular a series of adjoint equations to provide the value of an adjoint variable μ. The value of the adjoint variable is then used to determine which of five control modes to be employed: drive, hold, coast, regen or brake.
After describing the operation of the system taking into account inclines on the track, the specification summarises on page 11 lines 4-8:
“As a result, the optimal journey profile comprises driving in a hold mode (i.e. at constant speed), calculated by the Pontryagin Principle of optimal control, wherever possible and where it is not possible changing as quickly as is safely possible at exactly the right location, again calculated by the Pontryagin Principle of optimal control, to drive (i.e. full power), coast or brake modes as necessary.”
Pages 11-13 continue to say that advice is generated and displayed by the system for a train operator to keep the train close to the optimal profile. In the preferred embodiment multiple preferred profiles are generated and the train is monitored via GPS to keep track of the speed and position of the train. Advice is given with consideration of the speed and position and the time remaining until the train is due at the next key location, with these factors used to select the most appropriate precomputed profile. The specification then concludes with examples of the system in operation on pages 13-18.
The 44 claims of the application are as follows, with claims 1, 15 and 29 as independent claims:
1.A method of monitoring the progress of a train on a rail network and providing driving advice in real time to an operator of said train, said method comprising:
i.estimating or determining parameters of said train;
ii.determining, by an optimal control algorithm employing an adjoint variable, an optimal journey profile for a journey from said train’s currently location to a target location that results in said train arriving at said target location as close as possible to a desired time and with minimum energy usage; said optimal journey profile including a speed profile for the train, sequence of discrete control modes for said train, and associated switching points between said control modes;
said optimal journey profile being determined by solving a system of differential equations for said speed profile of the train and for the value of said adjoint variable, said control modes being determined from the value of said adjoint variable, such that said sequence of control modes is determined as said speed profile is calculated;
iii.monitoring the current state of said train as it progresses to said target location; and
iv.generating said driving advice for the train operator by comparing the current state of the train to a corresponding state on said optimal journey profile and displaying said advice for the train operator that will keep the train close to said optimal journey profile.
2.The method as claimed in claim 1 wherein said discrete control modes for said train include drive, hold, coast and brake modes.
3.The method of monitoring the progress of a train on a rail network as claimed in claims 1 or 2 wherein said adjoint variable evolves according to a differential equation along with the position and speed of the train.
4.The method of monitoring the progress of a train on a rail network as claimed in claims 1 or 2 wherein the value of the adjoint variable is calculated directly from the speed of the train.
5.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 1 to 4 wherein a numerical method is used to solve said system of differential equations for said speed profile of the train and for the value of said adjoint variable.
6.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 1 to 5 wherein steps (i) to (iv) are performed as required so that said driving advice automatically adjusts to compensate for any operational disturbances encountered by the train.
7.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 1 to 6 wherein said parameters include train mass and mass distribution.
8.The method of monitoring the progress of a train on a rail network as claimed in claim 7 wherein said parameters further include maximum tractive efforts and maximum braking effort as functions of speed.
9.The method of monitoring the progress of a train on a rail network as claimed in claims 7 and 8 wherein said parameters further include coefficient(s) of rolling resistance.
10.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 1 to 0 wherein said driving advice is generated and displayed by a computer located on the train.
11.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 1 to 10 wherein step (iii) involves processing data from a GPS unit and train controls to determine the location and speed of the train.
12.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 1 to 11 wherein said optimal journey profile specifies the time, speed and control at each location between the current train location and the next target location on the network.
13.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 1 to 12 wherein said optimal journey profile is precomputed.
14.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 1 to 12 wherein a plurality optimal journey profiles corresponding to different journey times are calculated and the profile that has an arrival time at the target location closest to the desired arrival time is selected.
15.A method of monitoring the progress of a train on a rail network and providing information on the progress of the train in real time to an operator of said train, said method comprising:
i.estimating or determining parameters of said train;
ii.determining, by an optimal control algorithm employing an adjoint variable, an optimal journey profile for a journey from said train’s current location to a target location that results in said train arriving at said target location as close as possible to a desired time and with minimum energy usage; said optimal journey profile including a speed profile for the train, sequence of discrete control modes for said train, and associated switching points between said control modes;
said optimal journey profile being determined by solving a system of differential equations for said speed profile of the train and for the value of said adjoint variable, said control modes being determined from the value of said adjoint variable, such that said sequence of control modes is determined as said speed profile is calculated;
iii.monitoring the current state of said train as it progresses to said target location; and
iv.generating said information for the train operator by comparing the current state of the train to a corresponding state on said optimal journey profile and displaying said information for the train operator to assist in keeping the train close to said optimal journey profile.
16.The method as claimed in claim 15 wherein said discrete control modes for said train include drive, hold, coast and brake modes.
17.The method of monitoring the progress of a train on a rail network as claimed in claims 15 or 16 wherein said adjoint variable evolves according to a differential equation along with the position and speed of the train.
18.The method of monitoring the progress of a train on a rail network as claimed in claims 15 or 16 wherein the value of the adjoint variable is calculated directly from speed of the train.
19.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 15 to 18 wherein a numerical method is used to solve said system of differential equations for said speed profile of the train and for the value of said adjoint variable.
20.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 15 to 19 wherein steps (i) to (iv) are performed as required so that said driving advice automatically adjusts to compensate for any operational disturbances encountered by the train.
21.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 15 to 20 wherein said parameters include train mass and mass distribution.
22.The method of monitoring the progress of a train on a rail network as claimed in claim 21 wherein said parameters further include maximum tractive efforts and maximum braking effort as functions of speed.
23.The method of monitoring the progress of a train on a rail network as claimed in claims 21 or 22 wherein said parameters further include coefficient(s) of rolling resistance.
24.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 15 to 23 wherein said information is generated and displayed by a computer located on the train.
25.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 15 to 24 wherein step (iii) involves processing data from a GPS unit and train controls to determine the location and speed of the train.
26.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 15 to 25 wherein said optimal journey profile specifies the time, speed and control at each location between the current train location and the next target location on the network.
27.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 15 to 26 wherein said optimal journey profile is precomputed.
28.The method of monitoring the progress of a train on a rail network as claimed in any one of claims 15 to 26 wherein a plurality of optimal journey profiles corresponding to different journey times are calculated and the profile that has an arrival time at the target location closest to the desired arrival time is selected.
29.A method of controlling the progress of a train on a rail network, said method comprising:
i.estimating or determining parameters of said train;
ii.determining, by an optimal control algorithm employing an adjoint variable, an optimal journey profile for a journey from said train’s current location to a target location that results in said train arriving at said target location as close as possible to a desired time and with minimum energy usage; said optimal journey profile including a speed profile for the train, sequence of discrete control modes for said train, and associated switching points between said control modes;
said optimal journey profile being determined by solving a system of differential equations for said speed profile of the train and for the value of said adjoint variable, said control modes being determined from the value of said adjoint variable, such that said sequence of control modes is determined as said speed profile is calculated;
iii.monitoring the current state of said train as it progresses to said target location; and
iv.comparing the current state of the train to a corresponding state on said optimal journey profile and then controlling said train to keep the train close to said optimal journey profile.
30.The method as claimed in claim 29 wherein said discrete control modes for said train include drive, hold, coast and brake modes.
31.The method of controlling the progress of a train on a rail network as claimed in claims 29 to 30 wherein said adjoint variable evolves according to a differential equation along with the position and speed of the train.
32.The method of controlling the progress of a train on a rail network as claimed in claims 29 or 30 wherein the value of the adjoint variable is calculated directly from speed of the train.
33.The method of controlling the progress of a train on a rail network as claimed in any one of claims 29 to 32 wherein a numerical method is used to solve said system of differential equations for said speed profile of the train and for the value of said adjoint variable.
34.The method of controlling the progress of a train on a rail network as claimed in any one of claims 29 to 33 wherein steps (i) to (iv) are performed as required so as to automatically adjust to compensate for any operational disturbances encountered by the train.
35.The method of controlling the progress of a train on a rail network as claimed in any one of claims 29 to 34 wherein said parameters include train mass and mass distribution.
36.The method of controlling the progress of a train on a rail network as claimed in claim 35 wherein said parameters further include maximum tractive efforts and maximum braking effort as functions of speed.
37.The method of controlling the progress of a train on a rail network as claimed in claims 35 or 36 wherein said parameters further include coefficient(s) of rolling resistance.
38.The method of controlling the progress of a train on a rail network as claimed in any one of claims 29 to 37 wherein step (iii) involves processing data from a GPS unit and train controls to determine the location and speed of the train.
39.The method of controlling the progress of a train on a rail network as claimed in any one of claims 29 to 38 wherein said optimal journey profile specifies the time, speed and control at each location between the current train location and the next target location on the network.
40.The method of controlling the progress of a train on a rail network as claimed in any one of claims 29 to 39 wherein said optimal journey profile is precomputed.
41.The method of controlling the progress of a train on a rail network as claimed in any one of claims 29 to 39 wherein a plurality of optimal journey profiles corresponding to different journey times are calculated and the profile that has an arrival time at the target location closest to the desired arrival time is selected.
42.A method of monitoring the progress of a train on a rail network and providing driving advice in real time to an operator of said train substantially as herein described with reference to any one of the embodiments of the invention illustrated in accompanying drawings and/or examples.
43.A method of monitoring the progress of a train on a rail network and providing information on the progress of the train in real time to an operator of said train substantially as herein described with reference to any one of the embodiments of the invention illustrated in the accompanying drawings and/or examples.
44.A method of controlling the progress of a train on a rail network substantially as herein described with reference to any one of the embodiments of the invention illustrated in the accompanying drawings and/or examples.
Evidence

Evidence in the form of declarations was filed as follows:
Evidence in Support
-          Anthony Howker dated 3 March 2011 (“Howker Declaration”).
-Professor Brian Anderson dated 17 May 2011 (“First Anderson Declaration”) with exhibits BA-1 to BA-6.
-Professor Brian Anderson dated 1 September 2011 (“Second Anderson Declaration”) with exhibits BA-7 to BA-13.
Evidence in Answer
-          Peter Pudney dated 4 October 2012 (“First Pudney Declaration”) with annexure PP-1.
-          Professor Kok Lay Teo dated 4 October 2012 (“Teo Declaration”) with annexure KLT-1.
-Dale Franklin Coleman dated 5 October 2012 (“Coleman Declaration”) with annexures DC-1 to DC-7.
-Russell James Davies dated 17 October 2012 (“Davies Declaration”) with annexure RJD-1.
Evidence in Reply
-Professor Brian Anderson dated 10 January 2014 (“Third Anderson Declaration”) with exhibits BA-1 to BA-6.
Further Evidence
-          Peter Pudney dated 16 June 2014 (“Second Pudney Declaration”).
The opponent noted that that Professor Anderson is an independent expert, as is Professor Teo. On the other hand they contend Dr Pudney has an interest in the proceedings as the inventor, and Mr Coleman has a commercial interest in the proceedings, and therefore are not independent experts. The opponent asserts that as per Bradken Resources Pty Ltd v Lynx Engineering Consultants Pty Ltd [2012] FCA 944 at 61, less weight should be given to non-independent experts.
I broadly agree with this principle. However, this is only one factor of many factors to consider, and only once all of these factors have been considered can I come to a position as to what weights I should or shouldn’t attribute to each piece of evidence. Such factors are further considered in my reasons below.
The Skilled Addressee

An important starting point is the determination of the skilled addressee, a non-inventive person working in the art who is armed with the common general knowledge of the art at the relevant date of priority. The nature of the skilled addressee was discussed in Root Quality Pty Ltd v Root Control Technologies Pty Ltd [2000] FCA 980 at [71]:
Generally speaking the skilled addressee is the person who works in the art or science with which the invention is connected. In Plimpton v Malcolmson (1876) 3 ChD 531 Jessel MR said (at 556):
"What is meant is that if [the invention] is a manufacture connected with a particular trade, the people in the trade shall know something about it; if it is a thing connected with a chemical invention, people conversant with chemistry shall know something about it."
In Catnic Lord Diplock said (at 242) that skilled addressees are "those likely to have a practical interest in the subject matter of [the] invention". A variety of people may have that interest. There are those who might wish to make or construct the invention, those who may wish to compound the invention and those who may wish to use the invention. The skilled addressee seems to me to be a relative expression which does not identify any specific person. Because the patent is directed to a person interested in making, constructing, compounding or using the invention (see eg s 27(3)(b) of the Patent Act 1993 (Canada); International Standard Electric Corporation v Ooms 157 F2d 73 (1946)), this hypothetical person, the patent lawyer's "reasonable man", may be required to be skilled in more than one art. Such a person might be thought of as the composite being, mentioned by Buckley LJ in Tetra Molectric, above, at 583. It may be preferable not to search for a composite addressee but a team whose combined skills are to be employed. In The General Tire & Rubber Company v The Firestone Tyre & Rubber Company Ltd [1972] RPC 457 at 485, Sachs LJ said:
"The construction of these documents is a function of the court, being a matter of law, but, since documents of this nature are almost certain to contain technical material, the court must, by evidence, be in the position of a person of the kind to whom the document is addressed, that is to say, a person skilled in the relevant art at the relevant date. If the art is one having a highly developed technology, the notional skilled reader to whom the document is addressed may not be a single person but a team, whose combined skills would normally be employed in that art in interpreting and carrying into effect instructions such as those which are contained in the document to be construed."
In their written submissions the opponent has submitted that the “background of the specification describes the problem in a form such that it is directed to be solved by a person skilled in the theory and practical application of optimal control.” They submit that Professor Anderson is well qualified to solve such a problem, as he has a background in electrical engineering, mathematics and train control, as well as extensive experience in working on optimal control problems on other physical systems such as aircraft control.
In contrast, the applicant says that they agree with the definition of the person skilled in the art as provided by the opponent in the statement of grounds and particulars, namely “a person or team of people working in the field of the operation and control of trains on a rail network.” They submit that this would include an engineer responsible for the operation of trains, but not necessarily an optimal control expert. Taking this further, the applicant alleges that there is no evidence on whether such a person skilled in the art would have knowledge of optimal control, nor if they would include an optimal control expert in any team (if such a team were even formed) to confront the problem. It then follows from the same argument that Professor Anderson should not be considered to be a person skilled in the art.
I agree with the applicant in that the problem is one of minimising energy usage for a train journey. The specification discusses the problem in the context of principles of efficient train driving and other existing methodologies which “may not result in the most energy-efficient journey”[1]. I also agree with the applicant that this problem is likely to be faced by someone working in the train industry, and as such these people which would include train engineers and railway operators.
[1] Page 2 lines 11-12 of the description.
In the event that I was to agree with the opponent that a person skilled in the art would be a person skilled in the theory and practical application of optimal control, the evidence suggests that train engineers and railway operators would also be included in a team alongside the optimal control specialist. Professor Anderson’s past work experience includes projects such as control algorithms for unmanned airborne vehicles and simplifying the design process for the pitch control system employed in commercial aircraft. His experience with train control is provided at paragraph 6 of the First Anderson Declaration which describes Professor Anderson attending a presentation on trains controlled on lines connecting mines with ports. It is clear to me that although Professor Anderson is well qualified to solve problems involving optimal control, besides attending a single presentation he has not worked with rail technology at all. More generally, an optimal control specialist such as Professor Anderson is more likely to operate at a high level of approach to general optimal control problems, and when tackling specific problems on a physical system this optimal control specialist would require the help of a domain specialist (such as Professor Anderson describes at paragraph 15 of the First Anderson Declaration). The domain specialist in this instance would be the aforementioned train engineers or railway operators who would be able to assist an optimal control specialist.
It appears that in all cases the person skilled in the art includes these train engineers and railway operators, whether by themselves in their own capacity or when brought in by an optimal control specialist as domain specialists. Specifically, train engineers and railway operators would clearly have a practical interest in the subject matter of the invention as it would directly affect their regular railway operations. The question now turns to whether an optimal control specialist should also be regarded as a person skilled in the art.
It is apparent from the specification that it is possible to solve the railway energy-consumption problem by using methods of optimal control as described, and as such a person skilled in optimal control theory would be able to solve the problem using their skills and knowledge. However I do not consider that they would be the person confronted with that problem initially. Rather, I would assert that an optimal control specialist would be more accustomed to facing optimal control problems over a number of fields and domains, rather than being restricted to facing problems exclusively within the field of train control. This certainly fits with Professor Anderson’s wider experience across a number of physical domains, and Professor Teo also has a similar range of experience. Therefore whilst an optimal control theory specialist can come across this problem by happenstance, it is more likely that way that they would come across this problem would be if they were sought out by a train engineer or railway operator, who as established earlier would be confronted with this problem on a much more regular basis through their work. And because the optimal control specialist is not someone who would work in the railway industry and be directly confronted with this problem, it is therefore more difficult to conclude that they are the person who would have a practical interest in the subject matter of the invention as opposed to just a passing theoretical interest in the subject matter.
However, this does not automatically exclude optimal control specialists from also being included within the definition of the skilled addressee as part of a team of people. I note the following passage from Minnesota Mining & Manufacturing Company v Tyco Electronics Pty Ltd [2002] FCAFC 315 at [91]:
The team approach is, of course, designed to solve a problem which requires the joint efforts of experts in more than one field. Such an approach must involve interaction between or amongst the experts so that common general knowledge can be shared. In this case the evidence did not deal with the way in which this process might have occurred.
(Emphasis added)
It is entirely reasonable that a person who works with trains, confronted with the problem of minimising energy usage and costs, would then explore avenues to solve this problem. It may also be very reasonable that a train engineer or operator would seek out the help of an optimal control specialist, and together they would form a team that would be able to solve the problem. However, for the present opposition proceedings I have no evidence from such a train engineer or railway operator that they would indeed go and seek out the assistance of an optimal control specialist, and even if such assistance was sought I have no evidence about how such a team might interact with each other to work towards a solution. In fact, the evidence provided by Professor Teo suggests that there are multiple options that might be available to solve the problem, as discussed at paragraph 5 of his declaration:
“The focus of Professor Anderson’s declaration is how he would solve the Train Control Problem using optimal control theory. However, it can also be solved by using optimization techniques through discretization of both the control and state or just discretization of the control.”
Dr Pudney similarly states at paragraph 55 of the First Pudney Declaration that:
“Accordingly, as at 2002, if a person was attempting to solve the train control problem, a range of potential approaches were available, of which the application of optimal control theory was just one.”
Professor Anderson’s comments on the same matter are found in the Third Anderson Declaration at paragraph 16:
“In paragraph 55, Dr Pudney asserts that optimal control theory is just one approach to minimising energy while maintaining schedules. That contention is true. However, in thinking about how one would seek to minimise energy, optimal control would have to be seen as a prime candidate for investigation.”
The consensus reached by each of the experts above is that there are multiple avenues available to be explored when seeking a solution to the problem. As noted before, I consider that it would be a train engineer or operator that would be the persons initially confronted with the problem. I have no evidence before me from such a person to suggest that optimal control would be the particular path that they would choose to go down or a path that they would even contemplate as even a possibility of going down. Even though Professor Anderson has contended that optimal control would be a “prime candidate for investigation”, he has not provided such a statement from the position as a train engineer or operator, but rather from his own experience of applying optimal control over the course of many years. The absence of such evidence does not assist the opponent’s submissions that the optimal control specialist should be considered to be a skilled addressee. Furthermore, even though the solution presented in the specification uses optimal control I am not satisfied that any characterisation of the problem as being an optimal control problem (as submitted by the opponent) is correct in light of these other options available – to pigeonhole the problem in such a way would be an impermissible use of hindsight.
One submission made by the opponent during the hearing was to analogise the problem to that of a patient and a doctor. It was argued that even if a patient is the one who is confronted with an initial problem of illness, they would seek out the help of a doctor, and it would be the doctor who would be considered as the person skilled to help the patient with his or her illness. I do not think that the presented analogy applies to the situation at hand. In the analogy it is assumed that the patient would know that when confronted with the problem (i.e. being sick), they should consult someone for assistance, and it is further assumed that the person knows they should seek assistance from is a doctor. In the present situation no such evidence has been put forward to support or validate these similar assumptions.
There are some final points worth addressing on this matter. The first relates to the language of the specification itself. The specification is to be read by the “skilled reader”, and although the specification provides details of the optimal control algorithm used, I am not of the opinion that the specification has been drafted in such a manner that only specialists in optimal control would be able to read and understand the specification. The other point raised as a general proposition by the opponent is the ability of the hearing officer to draw inferences on matters including those relating to the skilled addressee, as per Commissioner of Patents v Emperor Sports Pty Ltd [2006] FCAFC 26 at 24:
The Commissioner is an administrative decision-maker equipped with technical expertise. Subject to the rules of natural justice both common law and statutory…he or she is entitled to make use of that expertise, and draw inferences that may be rationally drawn from such technical knowledge, including how skilled persons of various descriptions may act in their respective occupations.
I have noted above that there is no evidence presented from a train engineer or operator regarding if and how they would employ the services of an optimal control specialist, and in this instance without any such evidence before me I am not prepared to make any assumptions that would otherwise fill this void. Therefore to reiterate, using the evidence before me, I consider that the person skilled in the art is a train engineer or a railway operator. This then has the effect that of the expert witnesses that have provided declarations in evidence before me, only Mr Howker and Mr Coleman have backgrounds in railway engineering, and can be considered persons skilled in the art for the current proceedings. In contrast, Professor Anderson has only had limited exposure to railway operation and control, and similar logic applies to Professor Teo and Dr Pudney. Nevertheless, the evidence presented by Professor Anderson, Professor Teo and Dr Pudney will still be of assistance for certain aspects such as understanding the operation of the present invention as well as providing technical insight into the operation of disclosures of the cited prior art documents.
Construction
The well known principles to be applied when approaching questions of construction are set out in Pfizer Overseas Pharmaceuticals v Eli Lilly and Company [2005] FCAFC 224 at [249], citing Sheppard J in Décor Corp Pty Ltd v Dart Industries Inc (1988) 13 IPR 385.
There are a number of terms found within the language of the claims which are worth examining in more detail. These features are discussed below.

“In real time”

Claim 1 recites that the method provides “driving advice in real time to an operator of a train”. Independent claim 15 also has a similar feature for the provision of information in real time. In both of these claims, the advice or information is being provided to the operator during the train journey, and the particular advice or information given allows the driver of the train to operate it more efficiently by adhering to the journey profile. Therefore “real time” should not be narrowly construed to mean that the advice is given instantaneously – rather the advice must be given at such a time that would allow the operator to adapt the operation of the train to keep to the journey profile. The claims also include within their scope (although are not limited to) “long-haul” train journeys, and so the scope of the term “real-time” would also necessarily include times which may extend slightly longer than those times afforded to metropolitan railway networks as the longer distances would allow for more leeway for decision making. Nevertheless the advice must still be given at a time allowing the train operator to alter the drive status of the train to realign with the optimal journey profile.
I further note that if advice or information is given in real time then this would exclude situations where advice was given to the operator via a manual or some other static source of information. The term “real-time” connotes that the advice given must have some temporal connection with the train journey.

“Estimating or determining parameters of said train”

On page 12 of their written submissions the opponent has stated:
“It is open for the person skilled in the art to choose, as a matter of routine, any suitable parameters according to the broadest claim.”
I disagree with this assertion. In addition to the parameters chosen being suitable, a person skilled in the art would also recognise that a sufficient number of parameters would need to be chosen in order for the system to produce a useful and useable result. As per the evidence[2] selecting too few would result in a system of no value, and choosing too many could result in an increase of complexity in the system of differential equations such that they cannot be solved.

“Generating said driving advice for the train operator by comparing the current state of the train to a corresponding state on said optimal journey profile and displaying said advice for the train operator that will keep the train close to said optimal journey profile”

[2] See for example Second Pudney Declaration at paragraph 8

The opponent has noted in their written submissions at paragraph 50 that the system of claim 1 provides driver advice which can be implemented at the operator’s discretion, and does not include “a closed loop control system in which the optimal control is used automatically to control the train (e.g. without input from the train operator)”.

I broadly agree with the above construction that claim 1 does not include systems which involve the automatic control of the train. However I also note that independent claim 29, which is directed towards a method of controlling the progress of a train on a rail network, does not include the feature of providing advice to a train operator, and instead simply recites:
“(iv) comparing the current state of the train to a corresponding state on said optimal journey profile and then controlling said train to keep the train close to said optimal journey profile.”
This means that claim 29 includes within its scope any trains which are controlled in an automatic fashion without the intervention of a human operator. In a similar fashion claim 29 does not require the presence of a display device, which is a feature present in independent claims 1 and 15.

“Wherein the value of the adjoint variable is calculated directly from the speed of the train”

This feature is found in dependent claims 4, 18 and 32. The word “directly” only places a limitation on the claims in so far as it requires that the speed of the train to affect the result of the co-state equation to obtain the value of the adjoint variable.
Novelty

Relevant Law

The test for determining novelty is the well-known “reverse infringement test” as set out by Aickin J in Meyers Taylor Pty Ltd v Vicarr Industries Ltd [1977] HCA 19 at [20]; [1977] HCA 19; (1977) 137 CLR 228 at 235:
“The basic test for anticipation or want of novelty is the same as that for infringement and generally one can properly ask oneself whether the alleged anticipation would, if the patent were valid, constitute an infringement.”

When considering prior art for novelty purposes, the prior art must contain “clear and unmistakable directions to do what the patentee claims to have invented” as per The General Tire & Rubber Company v The Firestone Tyre and Rubber Company Limited [1972] RPC 457 at 486; (1971) 1A IPR 121 at 138. The level of disclosure required from prior publications was also discussed in Nicaro Holdings Pty Ltd v Martin Engineering Co (1990) 91 ALR at 517:

It is well accepted that the prior art must disclose all features of the invention embodied in the patent in suit and must do so in clear, unequivocal and unmistakeable terms. The prior art must enable the notionally skilled addressee at once to perceive and understand and be able practically to apply the discovery without the necessity of making further experiments. Whatever is essential to the invention must be read out of or gleaned from the prior publication.

There may be instances where certain features are inherently disclosed in the prior art documents. Inherency was discussed in Novozymes A/S v Danisco A/S [2013] FCAFC6, firstly at 107:
The appellants relied on the judgment of Black CJ and Lehane J in Bristol-Myers Squibb Co v FH Faulding & Co Ltd [2000] FCA 316; (2000) 97 FCR 524, in which their Honours considered the leading authorities on the question of what was required in a prior publication to constitute anticipation, and continued (97 FCR at 548 [67]):
What all those authorities contemplate, in our view, is that a prior publication, if it is to destroy novelty, must give a direction or make a recommendation or suggestion which will result, if the skilled reader follows it, in the claimed invention. A direction, recommendation or suggestion may often, of course, be implicit in what is described and commonly the only question may be whether the publication describes with sufficient clarity the claimed invention or, in the case of a combination, each integer of it.
Disclosures in the prior art documents must also be enabling, which is discussed in Acme Bedstead Co Ltd v Newlands Bros Ltd [1937] HCA 63; (1937) 58 CLR 689 at 707 as being a:
…well-settled rule that a prior paper publication, giving information that does not become part of the common general knowledge, does not invalidate a subsequent patent unless it supplies enough information to enable a person of proper skill in the art to produce the mechanical device or appliance or carry out the process claimed in the later specification.

Khmelnitsky

Although multiple documents were raised in the particulars, during the hearing the opponent confirmed that they would rely solely on Khmelnitsky paper[3] (hereinafter “Khmelnitsky”) for novelty purposes. The opponent asserts that all claims 1-44 are anticipated by Khmelnitsky.

Claims 1-28

[3] “On an optimal control problem of train operation”, E. Khmelnitsky, IEEE Transactions on Automatic Control, Volume 45 Number 7, pp 1257-1266 (July 2000).

Unusually, although both the invention as disclosed in the current specification and Khmelnitsky discuss various aspects of mathematical equations and systems, the point of contention during the hearing between the claims and Khmelnitsky predominantly focused on the feature of whether the driving advice given to the train operator is displayed.
Khmelnitsky opens with the following paragraph:
One of the main aims of modern railroad engineering is the development of on-board locomotive computers and enlargement of their functionality. Applications of on-board computers allow management to act more swiftly in response to data from the field. This may be achieved by an advanced train control system (ATCS), which drives the train according to an optimal program and recalculates it when updated headway conditions appear. A necessary element of ATCS is an algorithm for finding an optimal program for traction and brake applications. The present paper proposes such an algorithm. It determines the velocity profile for minimizing energy consumption subject to the given data (trip time, grade profile, speed restrictions, and so on). In addition, the algorithm may also be used in training locomotive drivers.
It appears from this paragraph that the train which is the subject of Khmelnitsky is one which utilises an on-board computer which drives the train. Human intervention in the form of management appears to be done remotely, away from the train which is operating in “the field”. The only suggestion of the use of a human operator is provided in the final line of the paragraph through reference to the training of locomotive drivers.
The final paragraph of Khmelnitsky also makes reference to automatic control:
The problem of finding an optimal train operation which minimizes energy consumption is solved in this paper. An arbitrary variable grade profile and speed restrictions are taken into account. The maximum principle for the problems with mixed and state constraints is applied to obtain the analytical information about optimal operation regimes and their sequences. The presented numerical algorithm exploits this information to find the optimal velocity profile. Although the algorithm is purely open-loop, it can be incorporated in real-time feedback design in many different ways. For example, each time when the operator has to interfere in the train motion control (e.g., in the case of unexpected red lights or when the errors in the train position and/or speed exceed a given tolerance), the algorithm can restart and recalculate the optimal velocity profile by construction a new optimal program on the remaining trip length and time, using the current speed as the initial one. This recalculation can be carried out on an on-board locomotive computer for a few seconds, that is practically in real time. All of this allows us to expect that the proposed approach will significantly improve the present train operation technology.
(Emphasis added)
I have previously highlighted a submission by the opponent in relation to the construction of claim 1 that claim 1 does not include within its scope “a closed loop control system in which the optimal control is used automatically to control the train (e.g. without input from the train operator)”. It appears that in the context of the submission the opponent is equating closed-loop systems to automatic control. This is reinforced by the use of the term by Professor Anderson in the First Anderson Declaration at paragraph 25 where he describes automatic control such as the autopilot on an aircraft as “closed loop”. In contrast the reference in the above paragraph of Khmelnitsky to an “open-loop” system does not pertain to control but simply to feedback – the system presented in Khmelnitsky normally operates in such a manner that it calculates the journey profile based on parameters known before the journey, however it can also be modified to incorporate real-time feedback as the journey progresses and recalculate the “optimal velocity profile” as needed according to this new information. In that sense, this could be described as a “closed-loop” system simply because it allows feedback to be recycled into the system for recalculation, but it does not necessarily provide any disclosure about the method of control (between automatic control by a computer or manual control by a human operator) used on the train.
This then means that the only disclosures of the interaction between the system of Khmelnitsky and human operators come from the initial opening paragraph, and the references in the final paragraph to an operator interfering with the train motion control in situations such as an unexpected red light or where the train is outside the bounds of the expected speed or position. It is not clear where this human operator is based or what their function is (unlike the reference in the first paragraph to the locomotive driver).
Furthermore, there is no disclosure in Khmelnitsky about how the system presents information to the human operator. The system may be used for training locomotive drivers but it is not clear if this is done by presenting information directly to the driver, or by comparing the actions of the driver to those actions mandated by the system in an ideal calculated journey. If information is presented to the driver, it is also not clear if this is done visually or by other means such as audio. Finally, if there is information provided to the driver, it is not clear what nature this information takes, for example if it provides a comparison between the current state of the train to the corresponding state on the optimal journey profile, or simply provides an instruction to the driver on what action he or she should take at any given time. I would therefore conclude that there is no explicit disclosure of the feature of displaying advice or information for the train operator as per claims 1 and 15.
It is clear that the system of Khmelnitsky provides for an on-board computer, so therefore it is reasonable to conclude that the system is capable of incorporating a display device for use by the train operator. However, I am also of the opinion that it is far from inevitable that such a display would be present and would operate in such a way as to provide driving advice or information on the state of the train compared with the calculated state of the train according to the optimal journey profile. There is no indication in the Khmelnitsky paper that this is even contemplated, with most of the emphasis focusing on an automatically controlled train, and therefore without any human involvement. It is only for those offshoot applications such as locomotive driver training and a closed loop system do we see any mention of human involvement; otherwise the emphasis is on an automatically controlled system. Therefore there is no implicit disclosure of the display of advice or information to train operators.
Therefore as there is no disclosure, explicit or implicit, of displaying advice for the train operator, claim 1 is novel against Khmelnitsky. For similar reasons, with no disclosure of displaying information for the train operator, claim 15 is also considered to be novel against Khmelnitsky. As a result, each of the appended claims 2-14 and 16-28 are also considered to be novel against Khmelnitsky.

Claims 29-41

As noted above when discussing construction, claim 29 differs from independent claims 1 and 15 in that it is directed towards the control of a train, with no corresponding feature for displaying advice or information to a train operator, or any mention of a train operator at all. It therefore includes within its scope trains which are controlled using an automatic system, such as through a computer system.
This therefore immediately negates the issues discussed above with relation to the disclosure of Khmelnitsky against the claims 1 and 15. Therefore further examination of the Khmelnitsky paper is warranted.
The abstract states that the paper discusses a way of creating an optimal solution to control a train in such a way that it minimises the energy consumption as the train moves along a given route for a given time. This is done by determining a “detailed program for traction and brake applications”. This is further elaborated on page 1257 column 2 which says that the algorithm uses “mixed control and state variable constraints”, and also uses the maximum principle to obtain a sequence of control regimes “and the positions of their changing over”.
The algorithm utilises a number of variables, parameters and given data, listed over pages 1257-1258. The opening paragraph of the paper states that “trip time, grade profile, speed restrictions and so on” are incorporated. Part II entitled “Statement of the Problem: The Maximum Principle” also lists a number of variables and parameters, including the current time and train position, the trip time and trip length, the kinetic energy of the train at a particular train position on the journey, and control variables such as tractive and retarding forces. With respect to mass, the paper makes an assumption that the train is a mathematical point with unit mass, and declines to go further in expressing the train mass distribution.
Three differential equations are then provided to represent train motion, time motion and energy consumption. At paragraph 8 of the Third Anderson Declaration Professor Anderson notes the similarity between these equations to those found within the present opposed specification, specifically that “a model using three equations, based on the same physical principles as those in the Opposed Application with distance as an independent variable, and with dependent variables related by similar expressions to those in the Opposed Application, can be found in the Khmelnitsky Paper”.
At paragraph 9 of the Third Anderson Declaration Professor Anderson notes that the five control modes identified in the present opposed specification “are identified in the Khmelnitsky Paper which also identifies the same conditions for determining them, albeit with inconsequential notational differences”. At paragraph 30 he discusses these notational differences, and although Professor Anderson is not a person skilled in the art for these proceedings, I also agree that these notational differences do not have a consequential bearing on these proceedings. The control modes are determined via the calculation of a co-state variable, equivalent to the adjoint variable referred to in the opposed specification.
The final point of significance is taken from the final paragraph of the paper. As noted above in relation to claims 1 and 15, the final paragraph appears to confirm that the system provided for in Khmelnitsky is an open-loop system, but also envisages a closed loop system which can incorporate feedback in real-time. In this version the optimal velocity profile, equivalent to the optimal journey profile as claimed, is recalculated within “a few seconds” taking into account the change in circumstances. The examples provided for the closed loop system also include detecting “when the errors in the train position and/or speed exceed a given tolerance”, which suggests that the train position and speed are monitored and compared to set thresholds.
It would therefore appear that Khmelnitsky discloses each of the features of claim 29. In their written[4] and oral submissions the applicant has provided a number of arguments to attempt to distinguish Khmelnitsky from the opposed application.
[4] Applicant submissions paragraph 85.
The first argument revolves around the lack of disclosure in Khmelnitsky of provision of driving advice to the train operator. This is not a relevant consideration for the purposes of claim 29 as this is not a feature of the claim and so should be dismissed.
The second argument revolves around the viability of the system proposed in Khmelnitsky, specifically that “Khmelnitsky has not been shown to inevitably result in the determination of an optimal journey profile due to potential failure of the algorithm as a result of a failure to deal with the instability of the adjoint equations”[5]. This argument appears to be directed towards whether or not Khmelnitsky provides an enabling disclosure. Paragraph 17 of the Second Pudney Declaration states:
The instability of the equation containing the adjoint variable refers to the fact that when calculating an optimal journey over a long section of track, a small inaccuracy in calculation at the initial point of the journey, or early on in the calculation, can result in a false trajectory for the adjoint variable which leads to much more significant inaccuracy as the journey proceeds. This can have a large impact on the points at which the adjoint variable will indicate a change in control and can very significantly affect the ability to obtain an accurate optimal journey profile.
[5] Ibid.
Dr Pudney continues in paragraph 18:
I believe that the reason that Khmelnitsky may not have encountered instability issues (although we do not know for sure that he did not), could be that his examples do not use realistic gradients and relate to relatively short portions of track … In reality, an Australian train track for which an optimal journey profile is to be calculated may have hundreds, or even thousands, of different gradients along its length and may stretch for hundreds of kilometres. It is in these circumstances that the stability of the adjoint equation can become a real issue.
There is no discussion of instability in Khmelnitsky. From the evidence above it also appears that Dr Pudney is not sure if such instability issues would have been encountered by Khmelnitsky, only that they could have possibly been encountered. Dr Pudney also asserts that instability issues are generally more apparent in longer train journeys, at least those longer than the examples provided in Khmelnitsky. However the claims of the opposed application do not include any features that restrict the length of the journey to “long” train journeys. The only references to long-haul train journeys are derived from the title of the specification and the description. The claims themselves only refer to the progress of a train on a rail network, with a target location and current location.
I also note that Khmelnitsky is a peer reviewed paper published by the IEEE. No further evidence, whether by subsequent papers critical of Khmelnitsky or any other form, has been provided by the applicant beyond the observations of Dr Pudney to dispute the efficacy of the system in Khmelnitsky to produce an optimal velocity profile.
For the above reasons the argument put forward by the applicant that there is no enabling disclosure of an optimal journey profile due to potential instability in Khmelnitsky should be dismissed.
The final argument submitted by the applicant with regard to novelty regards the provision of advice in real-time. Although I note that Khmelnitsky provides an explicit disclosure that calculations can be performed in real-time[6], the feature of real-time advice and information only pertains to claims 1 and 15, which I have already considered to be novel against Khmelnitsky. Claim 29 does not include any features regarding real-time control, and therefore in relation to this claim the argument becomes moot.
[6] See last paragraph of Khmelnitsky, p1266.

I therefore consider claim 29 to lack novelty against Khmelnitsky.
Turning now to the appended claims, I find that the following claims have their features disclosed by Khmelnitsky:
·Claim 30: page 1259 describes control modes including drive, hold, coast and brake.
·Claim 31: Khmelnitsky describes the incorporation of the position of the train in the co-state equation on page 1258. Kinetic energy is similarly incorporated, which is a function of mass and speed.
·Claim 32: The term “directly” is construed to include within its scope a way of calculating the adjoint variable where the speed of the train affects the value of the variable. In this instance the co-state equation uses kinetic energy as a variable which itself is a function of speed, and therefore speed has a direct bearing on the adjoint variable calculations.
·Claim 33: the use of a numerical method is disclosed at page 1257.
·Claim 34: page 1266 discusses a closed-loop embodiment allowing for recalculation after an operational disturbance.
·Claim 40: Khmelnitsky does not explicitly disclose that the optimal journey profile is precomputed, but this feature is inherently disclosed. The system is described as open-loop on page 1266, which means that feedback is not used once the journey has commenced. For the train to be controlled along the journey, especially at points where control modes are switched, the optimal journey profile must be known. Therefore it is implicit that the optimal journey profile is precomputed.
Conversely, I do not find that claims 35-39 and 41 lack novelty. With regard to claim 35 Khmelnitsky specifically declines to include any parameters on mass distribution, instead assuming that the train is a mathematical point with unit mass[7]. It should be noted that although the opposed specification also treats the train as a mathematical point with unit mass (page 7 lines 16-18), it also does at least provide for an embodiment where the train mass distribution is considered, albeit in a uniform manner (page 6 line 17). Claims 36-37 are appended to claim 35 and therefore are also considered to be novel.
[7] Khmelnitsky p1258.
Claim 38 is considered to be novel as there is no disclosure of the use of GPS in Khmelnitsky, and no suggestion that the train location is monitored in such a way.
With regard to claim 39, Khmelnitsky provides examples on page 1265 which plot out the speed and control along a train journey at particular distances. However, there is no disclosure of the time at which these events occur. Although the time could be calculated easily for each point through the distance and speed, Khmelnitsky does not provide clear and unmistakable directions that time is specified by the system at each location. Therefore claim 39 is novel.
Claim 41 is considered to be novel as Khmelnitsky does not disclose the calculation of multiple optimal journey profiles with different journey times, and a subsequent selection of one of these profiles.

Claims 42-44

Claims 42-44 are omnibus claims corresponding to similar inventions covered by independent claims 1, 15 and 29. I consider claims 42-43 to be novel as Khmelnitsky does not disclose providing advice or information to an operator of a train as discussed above. I consider claim 44 to lack novelty for the same reasons provided against claim 29.

Summary

In summary claims 29-34, 40 and 44 lack novelty against Khmelnitsky.
Inventive Step

Relevant Law

Section 7 of the Patents Act states that:
(2) For the purposes of this Act, an invention is taken to involve an inventive step when compared with the prior art base unless the invention would have been obvious to the person skilled in the relevant art in light of the common general knowledge as it existed in the patent area before the priority date of the relevant claim, whether that knowledge is considered separately or together with the information mentioned in subsection (3).
(3) The information for the purposes of subsection (2) is:
(a) any single piece of prior art information; or
(b) a combination of any 2 or more pieces of prior art information;
being information that the skilled person... could, before the priority date of the relevant claim, be reasonably expected to have ascertained, understood, regarded as relevant and, in the case of information mentioned in paragraph (b), combined as mentioned in that paragraph.

Common general knowledge was discussed in Minnesota Mining and Manufacturing Co v Beiersdorf (Australia) Ltd [1980] HCA 9; (1980) 144 CLR 253 at 292 per Aickin J:
“The notion of common general knowledge itself involves the use of that which is known or used by those in the relevant trade. It forms the background knowledge and experience which is available to all in the trade in considering the making of new products, or the making of improvements in old, and it must be treated as being used by an individual as a general body of knowledge.
Approaches to inventive step have been discussed numerously in case law. In Wellcome Foundation Ltd v VR Laboratories (Aust) Pty Ltd [1981] HCA 12; (1981) 148 CLR 262 at 286, Aickin J stated:
"The test is whether the hypothetical addressee faced with the same problem would have taken as a matter of routine whatever steps might have led from the prior art to the invention, whether they be the steps of the inventor or not."
In Aktiebolaget Hassle v Alphapharm Pty Ltd [2002] HCA 59; 56 IPR 129 at 142-143 the High Court further explored the above approach:
That way of approaching the matter has an affinity with the reformulation of the ‘Cripps question’ by Graham J in Olin Mathieson Chemical Corporation v Biorex Laboratories Ltd [1970] RPC 157. This court had been referred to Olin in the argument in Wellcome Foundation. Graham J had posed the question:
`Would the notional research group at the relevant date in all the circumstances directly be led as a matter of course to try [the claimed invention] in the expectation that it might well produce [the desired result]?'
That approach should be accepted.
The word “ascertained” in section 7(3) means “discovered or found out” as per Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (No 2) (2007) 235 CLR 173 at 132. The ascertainable prior art will also depend upon the skilled addressee. In Commissioner of Patents v Emperor Sports Pty Ltd [2006] FCAFC 26 at 32-33 the Full Court said that in high technology areas it would usually be assumed that the relevant skilled person would be familiar with major professional or academic journals and could reasonably be expected to consult them, and in these instances no evidence would be required, however where the identification of the relevant prior art is in dispute “it is necessary to have either evidence or, which amounts to the same thing, reliance by an administrative decision-maker of expertise appropriate to the office”.

Prior art base

The opponent seeks to establish that the relevant prior art base includes Khmelnitsky, as well as a publication known as “The Red Book”[8] which was authored by the inventors of the opposed application.

Khmelnitsky

[8] Philip G Howlett and Peter Pudney, Energy-Efficient Train Control (Advances in Industrial Control) (Springer London, 1995).

Turning first towards Khmelnitsky, this publication was introduced into evidence at the Evidence in Reply stage. At paragraph 5 of the Third Anderson Declaration, Professor Anderson states how he discovered Khmelnitsky:
During my review, I noticed on page 14 at line 5 of the Opposed Application, a reference to authors “Howlett and Khmelnitsky”. I had previously assumed Khmelnitsky to be affiliated with the inventors. When I conducted an internet search for “Khmelnitsky”, I quickly located publication in the year 2000 by E. Khmelnitsky (IEEE Transactions on Automatic Control, Volume 45, July 2000, pp. 1257 ff, entitled “On an optimal control problem of train operation”) (the “Khemelnitsky Paper”).
I have previously established that Professor Anderson is not a person skilled in the art for the present proceedings, and so the evidence provided on how he ascertained Khmelnitsky is of low weight. However, even if Professor Anderson was to be considered a person skilled in the art, the methodology presented in the above paragraph is not be sufficient to establish that Khmelnitsky would have been ascertained by a person skilled in the art at the date of priority. Professor Anderson has used a reference in the opposed application to the author Khmelnitsky as the basis for conducting a search for literature. Such an approach clearly raises issues of hindsight, as a person skilled in the art at the time of priority would not have the benefit of springboarding from opposed application to conduct a similar search.
Also of significance is that Khmelnitsky was published in the IEEE database “IEEE Transactions on Automatic Control”. Upon browsing the database, it appears as though this database is not focused on train control problems, but rather deals with multiple aspects of control engineering in a number of applications. There is no evidence that a train engineer or operator of a railway network could have been reasonably expected to have consulted such a database such that they could have discovered Khmelnitsky, nor is there evidence of what methodology they would have used to search this database or even the full IEEE databases that could result in the discovery of this document. Although I would be comfortable with inferring that the person skilled in the art could have been reasonably expected to consult literature of a technical nature, without any such evidence I am not prepared to conclude that the person skilled in the art could have been reasonably expected to have searched this particular database in such a way as to have ascertained Khmelnitsky. Therefore I consider that Khmelnitsky does not form part of the prior art base for the purposes of assessing inventive step.

The Red Book

The Red Book suffers from similar difficulties as those with regard to Khmelnitsky. The only evidence presented on whether the Red Book would be ascertained by a person skilled in the art is from Professor Anderson, who is not a person skilled in the art.
After being asked as to whether he would require further information and how he would seek out such information, Professor Anderson sets out his search process in paragraph 70 of the First Anderson Declaration:
In order to identify any relevant text books, I entered the phrase “automatic train control” into the search function of the Amazon online bookstore website at Amazon.com. This is a broad search and I would expect that a searching for books using that phrase would identify any general texts relevant to the train control problem.
He continues in paragraph 71 to say that the Red Book was listed as the third result from his search, which was performed on 13 May 2011.
As with Khmelnitsky, there are also problems with this approach to searching when establishing whether the Red Book could have been reasonably expected to have been ascertained. The search term phrase used was “automatic train control”. It is not clear to me why a person skilled in the art, when confronted with a problem of minimising energy usage for a train journey could have been reasonably expected to use this particular phrase in a search. As discussed with relation to determining the person skilled in the art, there are multiple options available which could lead to potential solutions for this problem, and there is no evidence that a person skilled in the art would choose to go down the path of optimal control over other paths. However, even if the person skilled in the art would choose to explore optimal control, I cannot fathom why they could have been reasonably expected to specifically search for “automatic” train control, which would appear only to exclude literature regarding manual control of trains by drivers and operators.
Therefore as with Khmelnitsky, with no evidence presented as to how a person skilled in the art would conduct their search or where they would conduct their search, and further problems with the searching methodology presented in evidence, I cannot find that the Red Book could have been reasonably expected to have been ascertained by a person skilled in the art.

Common General Knowledge

Despite the absence of prior art citations, the claims must still be tested for obviousness against the common general knowledge in the art as it existed at the time of priority. However, there are significant problems with the evidence presented by the opponent in this regard. The opponent has sought to rely heavily on Professor Anderson to establish the common general knowledge in Australia at the time of priority, but I have already concluded that Professor Anderson is not a person skilled in the art in regard to the present problem, and therefore his observations on what knowledge he would draw on to solve the problem and what was otherwise generally known in the art do not garner much weight.
Similarly there is no evidence to suggest that either Khmelnitsky or the Red Book were part of the common general knowledge in the art. Professor Anderson stated that he regularly consulted the IEEE Transactions on Automatic Control database at the priority date[9]. Nevertheless, Khmelnitsky was only located by Professor Anderson after performing an author search, and it does not appear that Professor Anderson knew about this document prior to the Third Anderson Declaration, or at least did not bring the document to the attention of opponent during the Evidence in Support stage. This far from establishes that Khmelnitsky was widely disseminated and known in the art.
[9] First Anderson Declaration, paragraph 21.
With regard to the Red Book, Professor Anderson has also admitted that he was not aware of this publication as of the priority date[10]. Once again this does not assist the notion that the Red Book was part of the common general knowledge as there is no other evidence provided to support the notion that a person skilled in the art would be well aware of the Red Book and its contents.
[10] First Anderson Declaration, paragraph 71.
Finally, there is no evidence from a person skilled in the art, or evidence suggesting that the person skilled in the art would be well versed in adopting the Pontryagin Minimum Principle or Principle of Optimality or any other similar method to reach the invention as claimed. As noted above, although the evidence presented by Professor Anderson and Professor Teo (who has also made reference to the Pontryagin Principle) make it clear that the Pontryagin Principle may well have been part of the common general knowledge of the art amongst specialists in optimal control, their evidence on the knowledge or abilities of the person skilled in the art for the purposes of these proceedings do not attract much weight.
Therefore without evidence that a person skilled in the art would have knowledge of optimal control systems and methods that would result in the construction of an optimal journey profile for use in either providing advice or information to an operator or used to control a train in a manner that would allow minimisation of energy usage, I find that all of claims 1-44 involve an inventive step.
Section 40: Fair Basis and Clarity
The test for fair basis is taken from Lockwood Security Products Pty Ltd v Doric Products Pty Ltd [2004] HCA 58 at [69]:
“… the question is whether there is a real and reasonably clear disclosure in the body of the specification of what is then claimed, so that the alleged invention as claimed is broadly, that is to say in a general sense, described in the body of the specification.”
The application of this test calls for a comparison of the claims with the matter described in the specification (Lockwood v Doric supra at [77]).
A claim will lack clarity if a third party cannot ascertain whether an act would fall within the scope of the claim (Monsanto Company v Commissioner of Patents (1974) 48 ALJR 59 per Stephen J. at 60). However, a claim does not lack clarity simply because it uses inexact language or is difficult to construe, as long as it provides a “workable standard” suitable to the intended use (Minnesota Mining & Manufacturing Co v Beiersdorf (Aust) Ltd [1980] HCA 9 at [46]; (1980) 144 CLR 253 at 274).
The main thrust of the argument put forward by the opponent with relation to both fair basis and clarity is that if the invention of the opposed application requires certain parameters and a particular combination of equations to be selected so that instability issues can be avoided and the invention is still workable and accurate, that these equations and parameters should be present in the claims. The opponent contends that since the claims do not include these details, the claims go beyond the scope of the disclosure and lack fair basis. The opponent also submits that for similar reasons the claims lack clarity as if it would not be routine to identify what parameters should be used, the scope of the claims is indeterminate.
I do not believe that either of the arguments above can be supported. As noted in construction, a person skilled in the art would recognise that estimating or determining parameters would involve selecting appropriate parameters and equations that would allow a solution to be formed which allows a reduction in energy usage but also having the train arrive close to a specified time. The claims are fairly based as the description provides for such a real and reasonably clear disclosure of the selection of such parameters and equations. Similarly the claims are clear as the scope of the claims can be determined.
Manner of Manufacture
Section 18(1)(a) of the Patents Act reads:
(1) Subject to subsection (2), an invention is a patentable invention for the purposes of a standard patent if the invention, so far as claimed in any claim:
(a) is a manner of manufacture within the meaning of section 6 of the Statute of Monopolies;
The seminal authority on manner of manufacture with relation to methods is National Research Development Corporation v Commissioner of Patents (1959) 102 CLR 252 (“NRDC”) where it was said at 275:
The point is that a process, to fall within the limits of patentability which the context of the Statute of Monopolies has supplied, must be one that offers some advantage which is material, in the sense that the process belongs to a useful art as distinct from a fine art ... that its value to the country is in the field of economic endeavour.

100. Further decisions have also sought to clarify the application of NRDC when dealing with computer implemented inventions. One of the more recent decisions Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150 adopted an approach focused on the substance of the invention, and stated at [106]:
The determination whether the claimed invention is truly “an artificially created state of affairs” in satisfaction of NRDC is made not by some mechanistic application of the criterion of artificiality or physical effect, but by an understanding of the claimed invention itself. The invention is to be understood as a matter of substance and not merely as a matter of form.

101. This discussion was continued at [115]:
The High Court (in NRDC at 277) spoke in terms of a separate result achieved by the claimed method that has its own economic utility consisting in the improvement. By this reasoning, the High Court directed attention to the subject matter to which the claimed method was directed, which needed to exhibit the required characteristics of a manner of manufacture to be patentable.

102. The invention in Research Affiliates was directed to the generation of an index for a portfolio of assets. The Full Federal Court considered that the invention was not a manner of manufacture, and at [117]-[118] reasoned:

In the context of the claim, the significance lies in the content of the data rather than any specific effect generated by the computer. The computer-implementation is an essential integer of the claimed process. That is, of course, important. It is of particular importance in the assessment of, for example, novelty and infringement. However, in examining whether a claimed invention is properly the subject of letters patent, it is necessary to look not only at the integers of that claimed invention but also at the substance of that invention.
The claimed method in this case clearly involves what may well be an inventive idea, but it is an abstract idea. The specification makes it apparent that any inventive step arises in the creation of the index as information and as a scheme. There is no suggestion in the specification or the claims that any part of the inventive step lies in the computer implementation. Rather, it is apparent that the scheme is merely implemented in a computer and a standard computer at that. It is no part of the claimed method that there is an improvement in what might broadly be called “computer technology”.

103. The Full Court decided that the implementation of a scheme or abstract idea, which itself would not be considered to be patentable subject matter, would not be rendered patentable subject matter through implementation on a computer. The substance of the invention is determined not as a matter of form but involves an analysis of the alleged inventive step compared to the prior art. This approach of examining the substance of the invention was also mentioned in D’Arcy v Myriad Genetics Inc [2015] HCA 35[11].
[11] D’Arcy v Myriad Genetics Inc [2015] HCA 35 at [6] and [94].
104. Turning to the present claims, claim 29 is for a method of controlling the progress of a train on a rail network. The end result of the method as claimed is “controlling said train to keep the train close to said optimal journey profile”. The control of the train is clearly a useful physical result in relation to a material or tangible entity.

105. The other claim sets provide more difficulty. Claims 1 and 15 are directed towards methods of monitoring the progress of a train on a rail network and providing advice or information to the operator of the train. After the optimal journey profile is determined, the driving advice or information is then displayed to the train operator so that the train can be controlled in a manner to keep the train close to the optimal journey profile. However, the claims do not provide any further detail as to whether the train is actually controlled or operated according to the advice or information – the advice and information is only characterised as being able to keep the train on target if the advice or information is followed.

106. The immediate consequence of this is that the only physical result of working the methods of claims 1 and 15 is the display of advice or information on a screen. Although through monitoring the current state of the train it would appear that physical input readings are taken (for example receiving and processing GPS signals), the other steps of the method are generally mathematical in nature. This results in the invention as being characterised as one which is based in the real world and is not merely an abstract scheme by using real world parameters and inputs, and so therefore these are different to the purely abstract scheme considered in Grant[12].

107. Nevertheless, the remaining steps of the method involve processing these parameters through a mathematical algorithm, and then simply displaying the result with no further physical effect. The bulk of the specification also appears to focus on the workings of this algorithm so that by solving the relevant equations an instruction can be produced for how the train should be controlled according to an optimal control profile.
108. So in light of the approach used in Research Affiliates, where does the substance of the invention as claimed in claims 1 and 15 lie? Methods of optimal control to produce an adjoint variable are known in the prior art through Khmelnitsky, and it has been established that claims 1 and 15 are novel against Khmelnitsky on the basis that it doesn’t disclose that advice or information is displayed to the train operator. The fact that the parameters used in the state equations are taken from the real-world aspects of the train and its characteristics cannot be used in this instance to provide the requisite artificial state of affairs. Focusing then on the advancement over the prior art, the mere presentation of advice or information is of an abstract and intangible nature, which isn’t absolved by the use of a physical display device in much the same way as it the computer didn’t overcome the unpatentability of the scheme in Research Affiliates. The presentation of instructions to the driver is a concept which does not amount to subject matter which is patentable. It is also apparent that some of the exceptions to non-patentability do not apply in this instance as the mathematical calculations do not provide for the improved use of the computer itself, as was the case in Re International Business Machines Corporation v Commissioner of Patents [1991] FCA 625 for an improved curve on a computer screen, nor is the advice or information presented in such a manner that it provides a material advantage such as the improved readability of a form in Moore Business Forms Application [1979] AOJP 2521. Therefore as the substance of the invention is directed towards matter which is not patentable, claims 1 and 15 are not considered to be a manner of manufacture and similar reasoning applies to appended claims 2-14, 16-28 and corresponding omnibus claims 42-43.
[12] Grant v Commissioner of Patents [2006] FCAFC 120.

109. My conclusion in relation to claims 1-28 and 42-43 might seem inconsistent with my conclusion in relation to claims 29-41 and 44, in the sense that if the substance of the invention is such that claims 1-28 and 42-43 are not a manner of manufacture then the same conclusion should apply to claim 29 and its appended claims. The difference arises from the requirement of the Act to consider inventions “so far as claimed in any claim”. The substance of claim 29 lies in the production of a useful and tangible result as the invention defined by the claim clearly sets out an application of the information produced by the algorithm through the control of the train, which is a manner of manufacture. In contrast the substance of claims 1 and 15 lies in the mere display of the information without the information being used or applied.
110. Therefore claims 1-28 and 42-43 are not a manner of manufacture under s18(1)(a) of the Patents Act.
Conclusion
111. The opposition is successful. Claims 29-34, 40 and 44 are considered to lack novelty, and claims 1-28 and 42-43 are not a manner of manufacture.

112. As the specification contains patentable subject matter, I allow the applicant 2 months from the date of this decision in which to propose suitable amendments to overcome the above findings.
Costs
113. It is usually the case that costs follow the event. I see no reason to vary that usual approach on this occasion. As the opposition has been successful I therefore award costs against the applicant according to Schedule 8.
Matthew Lee
Delegate of the Commissioner of Patents