DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Drawings
New corrected drawings in compliance with 37 CFR 1.121(d) are required in this application because there appears to be a typographical error in Fig. 9B. Fig. 9B depicts a “2nd subset of simulations (ego agent absent)”. The figure includes a note reading: “Environmental agent had to slow down (though less than in the 1st subset of simulations) to not hit ego vehicle”. Since the ego vehicle is absent in this subset of simulations, the environmental agent, EA1, would not have to slow down to avoid it. The phrase “to not hit ego vehicle” could simply be deleted. Or the closed parentheses could be moved so that the note reads: “Environmental agent had to slow down (though less than in the 1st subset of simulations to not hit ego vehicle)”. Alternatively, the phrase could read something like: “Environmental agent had to slow down (though less than in the 1st subset of simulations) to not hit the other environmental vehicle”. Appropriate argument or correction is respectfully required.
Applicant is advised to employ the services of a competent patent draftsperson outside the Office, as the U.S. Patent and Trademark Office no longer prepares new drawings. The corrected drawings are required in reply to the Office action to avoid abandonment of the application. The requirement for corrected drawings will not be held in abeyance.
Any structural detail that is essential for a proper understanding of the disclosed invention should be shown in the drawing. MPEP § 608.02(d). Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claims contain subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for pre-AIA the inventor(s), at the time the application was filed, had possession of the claimed invention.
Each claim will be discussed under its respective number:
1.
A method for a vehicle, the method comprising:
detecting an [environmental] agent in an environment of the [ego] vehicle;
determining a first candidate ego policy for the [ego] vehicle and a second candidate ego policy for the [ego] vehicle;
for the first candidate ego policy, predicting a first work for the agent;
for the second candidate ego policy, predicting a second work for the agent;
based [in part] on the first work and second work, selecting the first candidate ego policy as ego policy for the vehicle; and
operating the vehicle based on the selected ego policy.
Before getting to the reason for the rejection of claim 1, the examiner first needs to explain how he interprets the claim and the disclosure.
In the present disclosure, see Fig. 9A, which is a depiction of a “1st subset of simulations (ego agent present)” when the “Policy for ego vehicle [i.e., ego agent]: drive straight through intersection when light turns green”.
When the system runs this 1st subset of simulations a “Potential hazardous event (collision) [is] detected”. Then, the system will “calculate energy of collision” then “calculating success metric”. In this simulation the “Environmental agent had to slow down to not hit ego vehicle”. The system will “calculate work to slow down [the environmental agent]” and then “Use [that calculated amount of work] in calculating environmental agent goal metric.”
See Fig. 9B for a “2nd subset of simulations (ego agent absent)” in which “Environmental agent had to slow down (though less than in the 1st subset of simulations)”. The system will “calculate work to slow down”. The system will then “compare with work form 1st subset of simulations”, then “Calculate difference”, then “Calculating environmental agent goal metric”, then the system will: “aggregated with reward metric of ego agent and/or success metric to determine overall score”.
How does all this relate to present Fig. 8? At the top of Fig. 8 it has the title: “1st ego agent policy (e.g., merge into left lane)”. This ego agent policy is depicted at times t1 and tn in Fig. 7 (see the two frames labeled “1st subset of simulations” on the left side of Fig. 7). Conceptually it would not matter what the ego agent policy is, such as whether it is to change lanes, “make a right turn” as in Fig. 5, or some other policy. Any one of these policies could be the “1st ego agent policy” simulated in the “1st subset of simulations”.
For each ego agent policy there is a 1st subset of simulations and a 2nd subset of simulations. In some cases in the 2nd subset of simulations, the ego vehicle may be absent, but in other cases, the ego agent may be present. In the latter case, the 1st subset of simulations may simulate what would happen if the ego agent changes lanes, while the 2nd subset of simulations would simulate what would happen if the ego agent stayed in its existing lane. This is, in fact, illustrated in Fig. 4.
Yet this is not what claim 1 is referring to when reciting:
determining a first candidate ego policy for the [ego] vehicle and a second candidate ego policy for the [ego] vehicle;
That is because in the case illustrated in Fig. 4, there is actually only one ego policy. That policy is to “merge into left lane”. That policy is simulated in the 1st subset of simulations. What term should be given to what the ego vehicle does in the 2nd subset of simulations is a subject that will be addressed later in this discussion, but it is emphatically not the “second candidate ego policy for the vehicle” as recited in claim 1. The metrics for each ego agent policy are determined using a 1st and 2nd subset of simulations.
This gets confusing when looking at figures such as Fig. 4 in the present disclosure.
After the above two clauses, claim 1 recites:
based on the first work and second work, selecting the first candidate ego policy as ego policy for the vehicle; and
operating the vehicle based on the selected ego policy.
This seems to elide out a lot of the disclosed invention to the point that it actually loses written description and becomes new matter. Rather it seems that these clauses should read:
based [in part] on the first work and second work, selecting the first candidate ego policy as ego policy for the vehicle; and
operating the vehicle based on the selected ego policy.
That is because Fig. 8 and the rest of the disclosure does not really teach that just based on the first and second work, the system will select the best candidate ego policy and then execute it. The disclosure presents something much more involved than that.
In Fig. 8, once the work for the agent is found by running a 1st and 2nd subset of simulations (which compare “a first candidate ego policy” such as change lanes to another behavior, such as maintain current lane as shown in Fig. 4), the system actually calculates a number of “metrics” plural, such as “distance, energy, work, etc” as shown in Fig. 8. These are all compared and a “difference between metrics” is calculated. Then an “environmental goal metric based on the difference” is calculated. Several more metrics are determined, such as a “collision metric” and a “success metric”. Using all these metrics, an “overall score for 1st ego agent policy (e.g., 1st cost metric)” is determined. This “overall score for 1st ego agent policy” is compared against a “2nd score, 3rd score…etc.” which represents scores for a 2nd ego agent policy, a 3rd ego agent policy, etc, respectively. Then, based on this comparison, the system according to Fig. 8 will “select a policy for the agent based on the comparison”.
This policy is not simply “based on the first work and second work” as claim 1 recites. It is based on an “overall score” for each policy, which is itself an aggregate of the “success metric and reward metric”. Only buried down the chain is the work metric.
Regarding the 2nd ego agent policy, 3rd ego agent policy, etc., the disclosed system can generate variations in policy and run simulations based on them. It seems that the system can not only consider what might happen if the ego vehicle changes lanes (1st ego agent policy) versus maintains its current lane (other behavior), it can also consider what might happen if the ego vehicle changes lanes at a faster speed (2nd ego agent policy) verses maintains its current lane (other behavior), or what might happen if the ego vehicle changes lanes while an environmental agent tries to speed up, or an environmental agent desires to change lanes.
The system can’t control what the environmental agent will actually do, but perhaps if a horrific accident (high energy collision) is predicted if, for example, an ego agent changes lanes at the same time as an environmental agent speeds up, the system will determine that the cost is too high and decide against that ego agent policy.
In any case, the disclosed system can at least consider a 1st ego agent policy (such as change lanes) in 1st subset of simulations and compare that to a 2nd subset of simulations (in which the ego vehicle behavior is to maintain the lane). The system can also run a 2nd ego agent policy (such as change lanes while speeding up) in a 1st subset of simulations (of the 2nd ego agent policy) and compare that to a 2nd subset of simulations (such as maintain lane while speeding up). Scores can be assessed and compared, and a final policy for the ego vehicle selected.
With all that in mind, it seems that claim 1 is broader than the original disclosure unless it adds “in part” by reciting:
based [in part] on the first work and second work, selecting the first candidate ego policy as ego policy for the vehicle; and
operating the vehicle based on the selected ego policy.
Claim 1 therefore receives a 35 U.S.C 112(a) rejection for lacking written description. In this case, the claim is new matter.
But even if the “in part” phrase were added, this does not mean that present claim 1 is taught by the prior art. That is partly because the MPEP 2111, following the courts, requires examiners to read “a claim in light of the specification,” which it notes “is a quite different thing from ‘reading limitations of the specification into a claim.’” To determine if the claim is in the prior art, what the claim means and how it should be interpreted must be established.
To do that the examiner will make notes, both in bold inset into the claims, and after the claims. In the course of this discussion, the examiner will issue rejections, as well as explain his interpretation of the claims and disclosure.
First, however, note that the original disclosure does not use the phrase “candidate ego policy” as fond in claim 1. The disclosure refers to “a set of potential policies” that are simulated in order to select “a single potential policy from the set”. Figs. 3B and 8 recite: “select a policy for the agent based on the comparison”. But the claimed concept is clear in the original disclosure that the set of potential policies are a set of candidate polices. That whole reason why the system simulates them is because they are candidates for execution by the vehicle.
2.
Claim 2 recites the following, for which there is not rejection:
The method of Claim 1, wherein
the first work is an estimated energy expended by the agent in response to the vehicle implementing the first candidate ego policy, and
the second work is an estimated energy expended by the agent in response to the vehicle implementing the second candidate ego policy.
3.
Claim 3 recites the following:
The method of Claim 1, wherein the first work is a relative metric comparing:
estimated energy expended by the agent in a first scenario, in which the vehicle implements the first candidate ego policy (in the present disclosure, see Fig. 8 for “in the present disclosure, see Fig. 8 for “Compare metrics (e.g., distance, energy, work, etc.) from 1st subset of simulations with metrics from 2nd subset of simulations”. From there the system will “calculate difference between metrics” etc.); and
estimated energy expended by the agent in a second scenario, in which the vehicle does not implement the first candidate ego policy,
wherein the second scenario is separate from the first scenario.
Is claim 3 distinctly claimed and have written description? To determine that it is necessary to recall that claim 1 recites that:
for the first candidate ego policy, predicting a first work for the agent;
for the second candidate ego policy, predicting a second work for the agent;
Claim 3 now adds that “the first work is a relative metric”. This relative metric is found, according to claim 3, by:
comparing:
estimated energy expended by the agent in a first scenario, in which the vehicle implements the first candidate ego policy; and
estimated energy expended by the agent in a second scenario, in which the vehicle does not implement the first candidate ego policy,
wherein the second scenario is separate from the first scenario.
When the disclosure discusses “scenarios” it relates to Fig. 2. Paragraph 0051 states “…predict a set of future scenarios (e.g., perform a set of simulations)”. Paragraph 0062 teaches that “a set of future scenarios preferably refers to the predicted environment (e.g., surroundings) that the ego vehicle will encounter at one or more time points in the future.” Paragraph 0063 teaches that “the future scenarios represent at least positions associated with environmental agents…and optionally further any or all of: …any derived parameters (e.g., energy, work, etc.)”.
Based on these references, when the disclosure uses the word “scenario” it can mean at least the end point of a subset of simulations, such as the “scenario” shown in the bottom left of Fig. 4, which is at time tn. The word scenario also can include the energy or work any of the agents did over the course of time from time t1 to tn.
So when present claim 3 recites determining the “estimated energy expended by the agent in a first scenario, in which the vehicle implements the first candidate ego policy” that can broadly and reasonably mean estimate the energy expended by the agent when over the course of a 1st subset of simulations from time t1 to time tn when the ego vehicle implements a first candidate ego policy.
In the examiner’s view, there doesn’t seem to be much difference, if any, between the way claim 3 is worded and saying: determine estimated energy expended by the agent in a 1st subset of simulations in which the vehicle implements the first candidate ego policy. That’s because the energy expended during a 1st subset of simulations is the energy expended in a first scenario.
This raises the question of why even use the term scenario in the first place? Are claims that use it distinctly claimed? The question revolves around a particular difficulty which is best illustrated by considering Figs. 3A and 4. The difficulty is: what should the policy of the ego vehicle in the 2nd subset of simulations be called, especially if the policy there is really that the “ego agent [is] absent”? The title of Fig. 3A is: “1st ego agent policy: merge into left lane”. But that is really only the policy in the 1st subset of simulations. What should be the name of the policy for what the ego agent does in the 2nd subset of simulations, which is basically be absent? For discussion purposes, it will be called the “behavior” of the ego vehicle (or ego agent) in the 2nd subset of simulations.
Similarly, Fig. 4 is entitled “1st ego agent policy: merge into left lane”. But again there is really only one ego agent policy being simulated and that is the ego agent’s policy in the 1st subset of simulations. In the 2nd subset of simulations, the “ego agent maintains current policy such as driving straight”. The fact that only one ego agent policy is measured in a 1st and 2nd subset of simulations matters a lot because influences the interpretation going all the way back to claim 1.
When claim 1 recites
for the first candidate ego policy, predicting a first work for the agent;
for the second candidate ego policy, predicting a second work for the agent;
it sounds like it is referring to Fig. 3, for example, in which the “first candidate ego policy” in the claim would map to “ego agent present” and be related to the 1st subset of simulations, while the “second candidate ego policy” in the claim would map to “ego absent” (or maintain the lane, as shown in Fig. 4) and be related to the 2nd subset of simulations.
But in fact, this is not a correct interpretation of claim 1. That is because determining the “work” of the first candidate ego policy is actually performed through a comparison between the 1st and 2nd subset of simulations. As already discussed, the name of the policy that the ego agent executes in the 2nd subset of simulations is not that clear in the disclosure, but it is not, as recited in claim 1, “the second candidate ego policy”. When claim 2 refers to that second candidate ego policy it is referring to a whole other ego agent policy and whole other set of simulations, it is not referring to the right side of Fig. 3A, for example. That is why in Fig. 8, there is an arrow from the “2nd ego agent policy” pointing to a 1st and 2nd subset of simulations. That is why paragraph 0068 states that “for each ego agent policy, performing a 1st subset of simulations and a 2nd subset of simulations”. For each set of 1st and 2nd subset of simulations there is only one ego agent policy. Another 2nd ego agent policy requires another different 1st and 2nd subset of simulations.
So getting back to the claims, it makes sense that there can be a different energy expended by an environmental agent in a first scenario (basically a simulation of a 1st ego policy) and a second scenario (a 2nd ego agent policy). But the original disclosure, does not teach, as claim 3 does, comparing the work across these two scenarios. All that is compared across scenarios (which is to say, all that is compared across ego agent policies) is the overall score. That is what is shown in Fig. 8.
Therefore, claim 3 lacks written description. This is a 35 U.S.C. 112(a) rejection.
But does claim 3 have written description when it states “wherein the first work is a relative metric”. Fig. 8 recites: “Compare metrics (e.g., distance, energy, work, etc.) from 1st subset of simulations with metrics from 2nd subset of simulations”. Note that the focus on claim 3 is the work of the “agent” which is not the ego agent but an environmental agent. One easy way to understand this is in reference to Fig. 4. The environmental agent has to slow down in the 1st subset of simulations when the ego agent changes lanes (note the ego agents lanes from time t1 to tn) while the environmental agent does not have to slow down in the 2nd subset of simulations when the ego agent maintains its lane. The distance the environmental agent has to slow down is D1.
The system can also determine how much work it took for the ego agent to slow down, or more generally the difference in work between the first and second subsets of simulations. To do this the system compares the work expended in the 1st subset of simulations to the work expended in the 2nd subset of simulations. The system can then use this (along with many other metrics) to make decisions about what ego agent policy to execute.
Note that Fig. 4 can be confusing because the “second candidate ego policy” recited in claim 1 might be confused with the behavior of the ego agent in the 2nd subset of simulations of Fig. 4. But as previously discussed, what the ego agent does in the 2nd subset of simulations is not termed a “second candidate ego policy” in the original disclosure and should not be thought of that way. Rather, each single policy has a 1st and a 2nd subset of simulations.
Based on all this, is claim 3 clear and supported with written description when it states that “the first work is a relative metric”. If this had been written to focus on comparing the work in a 1st subset of simulations to the work in a 2nd subset of simulations that would have written description. Paragraph 0021 recites that “a metric used to assess the impact includes a difference in modified energy and/or work associated with the scene and its environmental agents, which is able to meaningfully reflect how much energy and work is required to be exerted by environmental agents in response to certain behaviors of the ego agent.”
So reasonably a work metric can be “a difference in…work”. This fits with Fig. 8 where it says “compare metrics (e.g.,…work) from 1st subset of simulations with metrics form 2nd subset of simulations” and then “calculate difference between metrics”. When paragraph 0021 says take “a difference in…work” that fits with Fig. 8 when it says “calculate difference between metrics”. Overall, there is an amount of work expended by an environmental agent in a 1st subset of simulations, and an amount of work expended by an environmental agent in a 2nd subset of simulations. According to Fig. 8 and paragraph 0021, the system make calculate a “difference” between them. It is in this way that “the first work is a relative metric,” as recited in claim 3.
If the claim 1 were misinterpreted so that where it says “for the first candidate ego policy, predicting a first work for the agent” and “for the second candidate ego policy, predicting a second work for the agent” a reader understood it as meaning that “the first work for the agent” is the work of the environmental agent in a 1st subset of simulations, and the “second work for the agent” is the work of the environmental agent in a 2nd subset of simulations,” how could the claim 3 claim that “the first work is a relative metric comparing…”? In other words, if claim 1 were misinterpreted to refer to the left and right sides of Fig. 4, for example, when referring to the first and second ego agent policies, how could the work of these two separate policies themselves be relative? Relative to what? Again, when claim 1 refers to a first and second ego agent policy, it is not referring to a 1st and 2nd subset of simulations. It is rather referring to essentially comparing a 1st subset of simulations with one policy against another 1st subset of simulations with another policy.
Now that all that is clear, and in consideration of the last sentence of the last paragraph, the question arises: why perform a 2nd set of simulations at all? Why does the system care how much work the environmental agent had to expend in the 1st subset of simulations in Fig. 4 as compared to the 2nd subset of simulations? Why measure that when the system is just going to run an entirely new set of 1st and 2nd simulations with another ego agent policy? If the environmental agent expended X amount of work in the 1st subset of simulations under a 1st ego agent policy, and expended Y amount of work in the 1st subset of simulations under a 2nd ego agent policy, wouldn’t comparing those two values produce, at least in part, the information the system could use in order to set an ego agent policy?
Whether such a system would work or not doesn’t really matter. That is not what is disclosed. What is disclosed is what is shown in Fig. 8 in which a 1st ego agent policy is set, such as “merge into the left lane” and then the work that is expended (for at least the environmental agent) is determined for both the 1st and 2nd subset of simulations. That is clear in Fig. 8. Then a difference between those is determined. Then an “overall score for the 1st ego agent policy” is eventually determined. Then the system runs additional sets of simulations using a 2nd ego agent policy, and ultimately generates an overall score for the second ego agent policy, and perhaps more ego agent policies. The system eventually compares those overall scores and selects a policy.
Although the system does not simply measure the work an environmental agent needs to expend in a 1st subset of simulations for a 1st ego agent policy to the work an environmental agent expends in a different 1st subset of simulations for a 2nd ego agent policy, the thought of that reveals something about the present disclosure. Namely, that it seems that in order to compare the relative work (itself a comparison) the environmental agent expends between a 1st and 2nd ego agent policy, it would seem that the behavior of the ego agent in the 2nd subset of simulations would have to be the same for both the 1st ego agent policy and the 2nd ego agent policy. Keep in mind that what the ego agent does in the 2nd subset of simulations is not a numbered policy. It appears to function as a kind of benchmark for comparison purposes to the 1st subset of simulations.
Yet in as much as it functions as a benchmark for comparison with the 1st subset of simulations, it seems likely that it functions as a benchmark across ego agent policies as well. Otherwise, what would be the value of comparing the relative work across ego agent policies? If the 1st ego agent policy was merge left, and that was compared in the 2nd subset of simulations to the behavior of maintaining the same lane, and then the 2nd ego agent policy was speed up and that was compared in the 2nd subset of simulations to the same behavior as in the 1st ego agent policy, a useful comparison of how that would effect the environmental agent could be obtained. But if the 1st ego agent policy was merged left, and that was compared in the 2nd subset of simulations to a behavior of maintaining the same lane, and then the 2nd ego agent policy was speed up and that was compared in the 2nd subset of simulations to the behavior of the ego agent being absent, would comparing the difference in work that the ego agent had to expend between the 1st and 2nd ego agent policies really be useful? It doesn’t seem like it. It seems like the behavior of the ego vehicle in the 2nd subset of simulations would need to be consistent for all ego agent policies. That inference, in fact, strengthens the allowance of a previous May Mobility application in series of applications because the argument there was at least partly that the ego agent could be absent in multiple simulations.
Overall, the disclosure teaches that there is a comparison between 1st and 2nd subsets of simulations in any given ego agent policy, and then there is a comparison across ego agent policies. That comparison is of an overall score. It is important to see from Fig. 8 that the comparison across ego agent policies is not a comparison of a work metric or even a difference between work metrics. Fig. 8 teaches comparing “overall” scores. These overall scores are individually influenced by the difference in work metrics (perhaps barely) but the comparison of this overall score should not be confused with the step of comparing a work metric from a 1st subset of simulations to a work metric of a 2nd subset of simulations and finding their difference.
4.
Claim 4 recites the following:
The method of Claim 3, wherein
the vehicle is absent in the second scenario (in the present disclosure, this can mean that when claim 3 recites that the ego vehicle “does not implement the first candidate ego policy” this does not mean that the ego vehicle maintains its current lane, as in the 2nd subset of simulations in Fig. 4, but actually that the ego vehicle is altogether absent, as in Fig. 3A.).
No specific rejection is given to this claim.
5.
The method of Claim 3, wherein
the second work is a relative metric comparing:
estimated energy expended by the agent in a third scenario, in which the vehicle implements the second candidate ego policy; and
estimated energy expended by the agent in the second scenario.
Claim 5 lacks written description and is new matter. This is a 35 USC 112(a) rejection. That is because, as has just been shown in the foregoing discussion, the first scenario in the disclosure is related to the 1st and 2nd subset of simulations of the 1st ego agent policy, while the second scenario is related to the 1st and 2nd subset of simulations of the 2nd ego agent policy.
Claim 5 introduces a “third scenario.” That much is fine. As shown in Fig. 8, there can be a 3rd ego agent policy with its own 1st and 2nd subset of simulations, which means there can be a third scenario. But claim 5 says that actually “a third scenario” is related to when “the vehicle implements the second candidate ego policy.” The ego agent policies do not remain the same, at least not typically. But even if they did there is no written description in the original disclosure for comparing energy related to one ego agent policy with energy related to another ego agent policy. The disclosed system does compare work or energy across entire sets of simulations like that. The system compares overall scores, not each metric that form the overall scores.
For examination purposes, the claim will be interpreted as written.
6.
The method of Claim 1, wherein
the first candidate ego policy is selected on the basis of the first work being less than the second work.
This lacks written description under 35 USC 112(a). A first candidate ego policy might be selected in part on the basis of the first work being less than the second work, but in fact there are probably scenarios in which the first work is not less than the second work and the first ego policy is selected anyways because the difference in slight and other factors end up swamping it. But even if that is not the case, the examiner still does not think that the first ego policy is selected on the basis of the first work being less than the second work. That is because the first ego policy is selected over the second ego policy when the overall score of the first ego policy is better, not necessarily always when the first work is better.
It is true that paragraph 0085 teaches that the system can consider “the environmental agent goal metric” which can be thought of as a way to pick an ego agent policy that avoids forcing environmental agents to slam on the brakes. If the ego agent wants to merge left (as a 1st ego agent policy), as shown in Fig. 4, but it requires the environmental agent in the target lane to brake or brake sharply the ego agent may just maintain the current lane (as a 2nd ego agent policy). But that decision is really based on an overall score. That overall score contains a lot of other considerations, such as a collision metric. And these can swamp the environmental agent work metric. Consider Fig. 9A. It may be that the ego agent needs to slam on the brakes to avoid a collision with EA2, which will in turn cause EA1 to slam on the brakes. That will require a larger amount of work than if the ego agent just crashed into EA2 thus allowing EA1 to have a little more space to brake. Would the disclosed system pick the collision just because the work expended by environmental agent EA1 is lower in that case then in the case of the near miss? Of course not, but that is what the claim implies. But even if the applicant wants to claim the cases in which the ego agent policy is selected when the work of EA1 is lower, that still does not mean that that is how the decision is made. The decision about the ego agent’s policy is made based on comparing overall scores.
7.
The method of Claim 1, wherein
the agent is within a plurality of agents detected in the environment, and
wherein selecting the first candidate ego policy comprises:
for each of the plurality of agents, predicting a respective first work value for the first candidate ego policy;
for each of the plurality of agents, predicting a respective second work value for the second candidate ego policy;
determining a first aggregate work by aggregating the respective first work values across the plurality of agents;
determining a second aggregate work by aggregating the respective second work values across the plurality of agents; and
selecting the first candidate ego policy based on a comparison of the first aggregate work to the second aggregate work.
Is there an “aggregate work” metric? Yes. Paragraph 0020 states that there is a third variation in the disclosure that can be “additional or alternative to those described above”. Paragraph 0021 adds that an example of this is a metric that assess “how much energy and work is required to be exerted by environmental agents in response to certain behaviors of the ego agent. These metrics can further by easily aggregated, assessed, compared, and/or otherwise efficiently used.” Paragraph 0022 gives another “example” of using distance, but that does not appear to necessarily have to be included in the example in paragraph 0021 involving work.
It is also true, as the claim states the “selecting the first candidate ego policy comprises” what is claim. That is true because “comprises” is open ended. There is much else that goes into selecting the ego agent policy, as seen in Fig. 8. No rejection is specifically given to this claim.
8.
The method of Claim 7, wherein
determining the first aggregate work comprises selecting a maximum work among the respective first work values of the plurality of agents.
This claim is supported by at least paragraph 0101.
9.
The method of Claim 1, further comprising
calculating an impact energy of a collision associated with the vehicle implementing the first candidate ego policy (in the present disclosure, see Fig. 9A for a potential collision being detected in a 1st subset of simulations. Then, the system will “calculate energy of collision” then “calculating success metric”),
wherein selecting the first candidate ego policy is based on the calculated impact energy.
No specific rejection.
10.
A method for a vehicle, the method comprising:
detecting a pair of agents in an environment of the [ego] vehicle;
for a first agent of the pair of agents, predicting a first work expended by the first agent;
for a second agent of the pair of agents, predicting a second work expended by the second agent;
aggregating the first work and second work into an aggregate work;
based on the aggregate work, determining an ego policy for the vehicle; and
operating the vehicle based on the ego policy.
The disclosed system does not determine an ego policy “based on the aggregate work”. There is much more to it than that. This is a written description rejection under 35 USC 112(a). For examination purposes, the claim will be interpreted as written.
11.
The method of Claim 10, wherein
the aggregate work is a maximum work selected from the first work and second work.
In the present disclosure, a reasonable interpretation of the first part of paragraph 0101 is that it teaches taking an aggregate of each environmental goal agent’s goa metric. Aggregating these generates “a maximum parameter among the tracked environmental agents.” But this does not state that the environmental goal metric is limited to the work metric, nor that the aggregate work is “a maximum work selected from the first work and the second work.” The examiner interprets that quoted phrase as meaning that the system picks whichever of the first work or second work that is higher. This is a 35 USC 112(a) rejection.
12.
The method of Claim 10, wherein
the work is predicted in units dimensionally equivalent to energy.
This is supported by paragraph 0021. No specific rejection.
13.
The method of Claim 10, wherein
predicting the first work comprises comparing work expended by the first agent in a scenario in which the [ego] vehicle implements the ego policy to a work expended by the first agent in a scenario in which the [ego] vehicle does not implement the ego policy.
This lacks written description. Thios is a 35 USC 112(a) rejection. A “first work” as claimed is related to a 1st ego agent policy. That first work can be compared to a second work that is related to a 2nd ego agent policy. Each scenario is related to a specific single ego agent policy. “First work” is not compared to “a work” in another scenario. Only overall scores are compared, as shown in Fig. 8.
It is possible that the applicant may think that the examiner is overlying on Fig. 8. The argument could be that Fig. 8 is non-limiting and that Fig. 2, for example, should also be referred to. The examiner has explored that line of thinking, and is open to considering arguments along these lines, but paragraph 0117 says that “S400 can optionally include comparing aggregated metrics (e.g., cost functions, as shown in Fig. 8) among different sets of simulations, where each of these sets of simulations corresponds to a particular proposed policy for the ego agent, such that an optimal policy can be selected for the ego agent to implement (e.g., policy associated with a minimum cost function.” In the examiner’s view, Fig. 2 and Fig. 8 are complimentary.
14.
The method of Claim 10, wherein
determining the aggregate work further comprises calculating a collision energy of a predicted collision.
Fig. 8 shows a work metric on the left branch and a “collision metric (e.g., based on energy of simulated collision…)” on the right branch. The two are not the same the aggregate work does not comprise collision energy. This is a 35 USC 112(a) rejection.
15.
The method of Claim 10, wherein
predicting the first work comprises estimating a goal for the first agent, wherein the first work quantifies an impedance of progress of the first agent towards the goal.
The “goal metric” and the “first work” are different metrics the latter does not comprise the former as claimed. This is a 35 USC 112(a) rejection.
16.
The method of Claim 10, further comprising:
based on a time since the [ego] vehicle has last moved, determining a scaling factor; and
determining a weighted first work by weighting the first work according to the scaling factor, wherein the aggregate work is based on the weighted first work.
This claim is related to paragraph 0099. The examiner does not see this being related to a work, however. Therefore, the claim lacks written description and is new matter. This is a 35 USC 112(a) rejection.
17.
The method of Claim 16, wherein
the scaling factor decreases over time.
Paragraph 0100 teaches that the scaling factor can change with time, but the examiner does not see where it says that the scaling factor decreases with time. Therefore, the claim lacks written description and is new matter. This is a 35 USC 112(a) rejection.
18.
The method of Claim 10, further comprising:
determining a proximity of the first agent to the [ego] vehicle;
based on the proximity, determining a scaling factor; and
determining a weighted first work by weighting the first work according to the scaling factor, wherein the aggregate work is based on the weighted first work.
This claim seems related to paragraph 0122, but that paragraph refers to a goal metric, not work. Rather than teaching that the work metric is weighted it teaches that t Therefore, the claim lacks written description and is new matter. This is a 35 USC 112(a) rejection. The examiner will consider explanations of this claim and others accompanied by a paragraph citation from the specification in case the applicant disagrees with these rejections.
19.
The method of Claim 10, further comprising:
determining a class of the first agent;
based on the class of the first agent, determining a scaling factor; and
determining a weighted first work by weighting the first work according to the scaling factor, wherein the aggregate work is based on the weighted first work.
Paragraph 0096 says that the goal metric can be scaled based on a class label. But a goal metric is not a work metric. Therefore, the claim lacks written description and is new matter. This is a 35 USC 112(a) rejection.
20.
The method of Claim 10, further comprising:
determining that the [ego] vehicle has a right-of-way over the first agent;
based on the right-of-way, determining a scaling factor; and
determining a weighted first work by weighting the first work according to the scaling factor, wherein the aggregate work is based on the weighted first work.
Paragraph 0099 says that a scaling factor can be determined based on a right-of-way. This seems related to the goal metric not the work metric. Therefore, the claim lacks written description and is new matter. This is a 35 USC 112(a) rejection.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 11 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor, or for pre-AIA the applicant regards as the invention.
11.
The method of Claim 10, wherein
the aggregate work is a maximum work selected from the first work and second work.
The phrase “aggregate work is a maximum work selected from the first work and second work” is unclear. That is a 35 USC 112(b) rejection. For examination purposes, the claim will be interpreted to mean that the system picks a maximum work for each environmental agent and sums them for each of the ego vehicle policies considered. But the claim could also be interpreted to mean that the system picks the single work metric of the environmental agent with the maximum work among all the environmental work agents related to a 1st ego agent policy and then adds that to the highest single work metric relate to a 2nd ego agent policy. Yet that interpretation would lack written description because the system does not add work values across ego agent policies because the point is to compare those polices and pick the best one.
Additional Art
The prior art made of record here, though not relied upon, is considered pertinent to the present disclosure.
One close prior art is Caldwell et al. (US2021/0046924), cited in a parent application. Caldwell teaches simulating various “potential vehicle actions,” scoring them based on a number of factors, and then determining the best action to take. Yet Caldwell does not teach make a score relative between a 1st ego agent policy and another ego agent behavior, determining a difference between metrics across those two, and then further doing that again for a 2nd ego agent policy and yet another ego agent behavior to obtain yet another difference between metrics, and then using those differences to generate an overall score, and then selecting the policy with the best overall score, as the present disclosure does.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DANIEL M. ROBERT whose telephone number is (571)270-5841. The examiner can normally be reached M-F 7:30-4:30 EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Hunter Lonsberry can be reached at 571-272-7298. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/DANIEL M. ROBERT/Primary Examiner, Art Unit 3665
