//
austlii:cth/APO/2025/22
General information, not legal advice.
APO
2025-07-15
Original judgment source is linked above.
Integer 1(ii): It is implicit from OS [Opponent submissions] that the Opponent accepts D1 does not disclose any particular type of accelerometer, let alone a three-axis accelerometer. The submission at OS [87] relies on the proposition that it was "inevitable" a PSA would select out of the myriad sensors described in D1 an accelerometer, and would then select a three-axis accelerometer, or an arrangement of three single-axis accelerometers, when implementing D1. That submission cannot be sustained.
The contention that it was CGK at the Priority Date to use three-axis accelerometers (or three single-axis accelerometers) for orientation detection is unsupported by probative evidence. In his first declaration, Dr Hiller provides a general account of sensors used in systems for monitoring GETs, including inertial sensing systems, but does not identify any instance in which three-axis (triaxial) accelerometers were known to be used in any such application, including in the context of wear element loss detection. While Dr Hillier asserts in his second declaration that the use of three-axis accelerometers for orientation detection was well-known at the Priority Date, he does not provide any specific references to technical publications or other contemporaneous materials to support that evidence which demonstrate such use at the Priority Date.
Integer 1(iii): Dr Vercellone asserts that "D1 makes no mention of selecting ... reference axes", and further, "D1 also does not direct me to use an accelerometer to determine the orientation of a GET". This is another example of Dr Vercellone adopting an unduly narrow and literal approach to disclosure. There can be no debate that D1 teaches the use of accelerometers in the monitoring device, nor that D1 teaches the use of the monitoring device to determine orientation of a GET. That being so, as Dr Hillier explained, "[d]efining orientation by reference to a reference axis would be immediately apparent as a possible approach to the PSA", "defining the orientation by reference to reference axes (about which the rotation is measured) would be the most likely approach the PSA would have taken" and "any orientation must be defined relative to some other (known) orientation and the reference is most commonly defined by an orthogonal set of (reference) axes" (emphasis added). Indeed, the reality is that any orientation must be defined relative to some other (known) orientation. Thus, the use of a reference would be "required as part of the way to define orientation", as taught in D1. Notably, Dr Vercellone does not suggest that the use of a reference axis would involve some contribution "beyond the skill of the calling" or any difficulty to be overcome. Rather, the extent of his evidence seems to be that he cannot locate a literal disclosure of this routine implementation step.
Integer 1(iii): The submissions at OS [89] rest on the assumption that, when following D1, the PSA would implement a system for monitoring attachment of a wear element using orientation measurements obtained from an accelerometer. As observed above, D1 refers to the possible inclusion of various types of sensors in the "monitoring device". It does not state that an "accelerometer" is to be used for orientation, as opposed to any other purpose. The submissions at OS [89] are therefore circular. They assume that an accelerometer would be used to determine orientation and then argue that, having made that assumption, it would be routine to define orientation with respect to a reference axis. For the reasons outlined above, the underlying proposition is not established on the evidence.
Integer 1(vi): Dr Vercellone asserts that "D1 does not tell me to use an accelerometer measurement to determine the rotational position of a GET". Again, this is an example of Dr Vercellone adopting an unduly narrow and literal approach to disclosure. The point being made by Dr Vercellone is that, even though D1 teaches the use of accelerometers in the monitoring device, and even though D1 teaches the use of the monitoring device to determine orientation of a GET, D1 does not expressly state "what characteristics an accelerometer is to determine" and it would also be possible to use an accelerometer to measure other characteristics (such as "the extent of usage of the GET"). This may be literally true, but is quite irrelevant. The question is whether it would be obvious to a PSA, armed with D1, which teaches the use of accelerometers and the measurement of rotational position of GET, to use accelerometers to measure rotational position of GET. In circumstances where Dr Vercellone has admitted that "[a]ccelerometers were known and used in a wide variety of devices and vehicles prior to July 2019", and the evidence makes plain that "it is, and was at July 2019, well known that accelerometers can be used for orientation detection", it would have been obvious to a PSA, armed with D1, to use the type of sensor disclosed (an accelerometer) to detect the metric disclosed (rotational position, that is, orientation).
Integer 1(vi): The submissions at OS [90] overlook the significance of the point being made by Dr Vercellone. In the absence of any express teaching in D1 regarding the use of an accelerometer for orientation detection, it is entirely appropriate to consider other potential uses of such a device. That observation is not irrelevant pedantry, but a reflection of the lack of disclosure in D1. Where a component has several plausible applications, it cannot simply be assumed that a PSA would understand a bare reference to that component as teaching its use for the purpose that aligns with the claimed invention. Further, the uses postulated by Dr Vercellone - calculation of use time or assessments of machine operation or efficiency by reference to information provided by an accelerometer based on movement measurements - are entirely consistent with the limited teaching in D1 regarding the use of accelerometers, namely, to provide "information regarding the operating conditions in which the ground engaging product is being used" (at [0116]).
Integers 1(vii) & 1(viii): These integers require the calculation of a "mean reference value" based on the "rotational position measurements" of the wear elements and a "deviation value" of a wear element based on the deviation between the "rotational position measurement" of that wear element and the "mean reference value".
Dr Vercellone disputes that there is any disclosure of the determination of a mean reference value, on the basis that the disclosure in D1 "is vague" and does not expressly recite that "there is a comparison made of one GET to another GET". It appears that Dr Vercellone denies the disclosure of a deviation value on the same basis.
Yet again, this involves Dr Vercellone requiring a strict, literal disclosure, rather than approaching the nature of the disclosure in D1, and how it would be implemented by a PSA at the Priority Date, with commonsense.
At [0017], D1 teaches that "Loss of a product member is detected when one monitoring device deviates beyond a certain range from its established position, or orientation." At [0122], D1 then further teaches that (emphasis added):
Loss of a product (e.g., through breakage or pin ejection) results in loss of the monitoring device from the established electronic framework; i.e., a significant change in position of the monitoring device is detected when one monitoring device deviates beyond a certain range of its established position with the other monitoring devices...Alternatively, the remote device 38 detects the positions of the monitoring devices 25 to ensure the presence of each product in its established position relative to the other monitoring devices 25. In either case, by detecting the presence of the product through its relative position and orientation with the other products, the system operates irrespective of the movement and operation of the earthworking equipment with the ground engaging products.
These passages make clear that D1 is teaching that a comparison be made between the rotational position of the monitoring device on a GET and the rotational position of the monitoring devices on the other GET (i.e. the "other monitoring devices" which are on "the other products"). As Dr Hillier has explained, "In order to detect the presence (or absence is implied) of the GET relative to the others via orientation, then evaluating the general or average orientation of the other GET is implicit". Dr Hillier also noted that his view of this point is supported by the "broad meaning of "mean reference value" that is established by the Opposed Application" and observed that when following the directions in D1 to compare the orientation of one GET to the other GETs, the PSA "would immediately consider comparing the orientation of the GET under investigation to an arithmetic mean of the orientation of the other GETs, as this is a very straightforward and intuitive way to determine how far one measurement deviates from a set of measurements".
Similarly, in order to determine whether a GET has "deviate[d] beyond a certain range" and its "relative position and orientation with the other products", it is necessary to undertake a comparison between the position/orientation of that GET and other GET, thereby resulting in the calculation of a deviation value. These are straightforward implementation details that a PSA would address when following the teaching of D1 to detect the orientation of a GET relative to the other GET, as described in D1.
Dr Vercellone merely asserts that "It is not necessary to obtain a mean reference value in order to compare the orientation of one GET with the orientation of another". However, that is not what D1 discloses. D1 at [0122] expressly, and repeatedly, teaches a comparison between the orientation of one GET "with the other monitoring devices", that is, all such other devices. The comparison which Dr Vercellone asserts to potentially avoid the need for a mean reference value is not the comparison which D1 teaches and discloses. It follows that the point made by Dr Vercellone goes nowhere.
Further, and importantly, Dr Vercellone does not anywhere suggest that the use or calculation of a mean reference value or deviation value would be "beyond the skill of the calling" or require any difficulty to be overcome. Indeed, it is to be recalled that the Application expressly admits that "the skilled person will understand that different formulations can be used to obtain representative values" and otherwise allows the PSA complete discretion as to how a mean reference value or deviation value is to be calculated, thereby revealing that this involves nothing more than the deployment of CGK. Thus, even taken at face value, the evidence of Dr Vercellone appears to rise no higher than an assertion that there might have been another, equally obvious, and technically straightforward, way of implementing the teaching of D1. That is no answer to an allegation of lack of inventive step.
Integer 1(ix): this integer requires a determination that there has been a fall of the wear element if its deviation value exceeds a threshold value. Although Dr Vercellone denies this is disclosed, that appears to be on the basis that D1 does not literally use the words "threshold value". It could not be clearer that D1 discloses this integer at [0017] & [0122], where it is explained that "significant change in position" of a monitoring device "beyond a certain range" will indicate "loss of a product". The "certain range" discussed in [0017] & [0122] is the "threshold value" required by this integer.
It follows that claim 1 of the Application does not involve an inventive step over D1 when combined with the CGK.
Integers 1(vii) and 1(viii): The submissions at OS [91]-[98] rely heavily on the proposition that D1 teaches or necessarily implies the calculation of a "mean reference value" based on "rotational position measurements" of the wear elements and a "deviation value" based on the deviation between the "rotational position measurement" of that wear element and the "mean reference value". That proposition cannot be sustained.
D1 does not disclose the use of comparative statistical operations - such as determining an average or deviation - as part of the detection method. Instead, it refers generally to detecting when a monitoring device "deviates beyond a certain range", without identifying what that range is or how it is derived. Dr Hillier's evidence regarding these integers - including that "evaluating the general or average orientation of the other GET is implicit" and that the PSA "would immediately consider comparing the orientation of the GET under investigation to an arithmetic mean of the orientation of the other GETs" - improperly elevates hindsight reasoning.
Further, the Opponent's attempt (at OS [97]-[98]) to dismiss Dr Vercellone's evidence is misplaced. Dr Vercellone correctly identifies that D1 does not describe determining a mean reference value based on the rotational position of all monitoring devices or GETs.
Regarding paragraph 49vii., Dr Vercellone says that D1 does not disclose the determination of a mean reference value based on rotational position measurements of each wear element. I disagree and stand by the comments in my First Declaration that, when following the directions at paragraph [0122], for example, to detect the presence (and therefore loss) of a product (GET) through its relative orientation with the other products (GETs), it is implicit that the orientation of each GET is compared to the general or average orientation of the others (described in D1 as the "established orientation"). My view on this point is strengthened by the broad meaning of "mean reference value" that is established by the Opposed Application. Regardless of how broad the meaning of "mean reference value" may be, when the PSA follows the directions in D1 to compare the orientation of one GET with the orientation of the other GETs, I believe they would immediately consider comparing the orientation of the GET under investigation to an arithmetic mean of the orientation of the other GETs, as this is a very straightforward and intuitive way to determine how far one measurement deviates from a set of measurements.
"...compare the orientation of one GET with the orientation of the other GETs..."
"...comparing the orientation of the GET under investigation to an arithmetic mean of the orientation of the other GETs..."
There are any multitude of ways to quantify that the orientation of one item of GET is different to the others - the most obvious being variations on that presented in AU '932, i.e. simple comparisons of the difference of the orientations of the GET against each other, and against a threshold, although I would consider minor adjustments in the method to take advantage of simple statistical outlier analysis methods would also be quite run-of-the-mill ways to identify the lost item of GET given this data.
The specific use of a weighted sum (or weighted mean) to get the reference value seemed unusual to me, but as I alluded to before, there are any number of ways that this concept can be implemented, and within reason, the various implementations would perform the same function without real impact to the overall method's or system's performance.
would each be an obvious step for a person skilled in the art. I have found that a person skilled in the art would not, as a matter of routine, modify D1 such that it includes the integers 1(vii), 1(viii) or 1(ix).
CQMS Pty Ltd
Metalogenia Research & Technologies S.L. [2025] APO 22