Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
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 .
Response to Arguments
Applicant's arguments filed 4/1/2026 have been fully considered but they are not persuasive. Applicant
Regarding 35 USC 101
35 U.S.C. & 101 Guidance - Step 2A:
Applicant submits that the amended claims are patent eligible under 35 U.S.C. § 101 for at least two reasons based on the 2019 Revised Patent Subject Matter Eligibility Guidance issued by the United States Patent and Trademark Office ("2019 Guidance"). See Manual of Patent Examining Procedure (MPEP) § 2106(l) (9th Ed., Rev. 10.2019, Last Revised Jun. 2020).
First, according to the 2019 Guidance, for a claim to constitute an abstract idea, the claim must recite limitations that incorporate mathematical concepts or constitute mental processes or certain methods or techniques of organizing human activity. See MPEP § 2106.04(a). Applicant submits that the amended claims do not recite any limitations falling within any of these enumerated groupings.
In that regard, the amended claims are not directed towards mental processes because the claimed steps cannot be practically performed in the human mind or using pen/paper. Claim limitations that encompass Al in a way that cannot be practically performed in the human mind do not fall within the mental process grouping. See Memorandum: Reminders on evaluating subject matter eligibility of claims under 35 U.S.C. 101 at 2 ("[a] claim does not recite a mental process when it contains limitation(s) that cannot practically be performed in the human mind, for instance when the human mind is not equipped to perform the claim limitation(s)"); see also MPEP § 2106.04(a)(2)(III). The amended claims recite the specific steps of processing the first prompt via a first trained language model to generate a second plan for driving the vehicle, and performing one or more operations to control the vehicle or generate output via an output device based on the second plan. These steps quite clearly require a computing device because processing a prompt using a trained language model is necessarily a computer-implemented step that cannot be performed entirely in the human mind or with pen/paper. Nor can a vehicle be controlled or an output be generated via an output device entirely in the human mind or with paper/paper. Because the amended claims recite steps that cannot be performed entirely in someone's mind or using pen/paper, the amended claims cannot be properly interpreted as involving mental steps.
The amended claims also do not recite any mathematical concepts. See MPEP 2106.04(a)(2)(I). In addition, the amended claims do not recite any approaches or techniques for organizing human activities, such as fundamental economic principles or practices, commercial or legal interactions, or personal behaviors or relationships or interactions between people. See MPEP § 2106.04(a)(2)(II).
Because none of the limitations recited in the amended claims are directed towards any of the enumerated categories of abstract ideas, the amended claims cannot be properly interpreted as being abstract.
Examiner respectfully disagrees. Examiner draws comparison to Claim 2 of Example 47 of the USPTO Subject Matter Eligibility Examples, published July 2024, which was found to be ineligible under 35 USC 101. The present claims, in summation, involve receiving information, generating a prompt requesting driving instructions based on said information, processing the prompt via a trained language model to generate a plan for driving the vehicle, and controlling the vehicle or generating an output based on the plan. Drawing comparison to Claim 2 of Example 47 of the Subject Matter Eligibility Guidelines, said Example Claim 2 involves processing a data set using an artificial neural network in order to generate anomaly data, which is subsequently output. In each case, the claims recite processing data using a model to generate an output, the model in each instance being a learning model. A learning model, under the broadest reasonable interpretation of the term, is a mathematical calculation, which falls under the mathematical concept category of abstract idea. While in the present claim, a “language” learning model (LLM) is recited, Examiner submits that the functioning of a LLM involves the conversion of language data into a mathematical representation, which is subsequently processed, rendering the use of an LLM a mathematical operation under the broadest reasonable interpretation of the claim. Further, even the requirement to use a computing element does not render a claim patent-eligible [MPEP 2106.04]. Thus, Examiner respectfully asserts that the amended claims are directed to an abstract idea of processing data using a learning model.
Second, the amended claims recite limitations that integrate any purported abstract idea into a practical application. See MPEP § 2106.04(d). In Ex Parte Desjardins, the Appeals Review Panel of the Patent Trial and Appeal Board explained that "[o]n the one hand, claims '[g]enerally linking the use of a judicial exception to a particular technological environment or field of use' are not patent eligible.... [o]n the other, claims directed to an improvement in the functioning of a computer, or an improvement to other technology or technical field are patent eligible." See Ex Parte Desjardins, Decision on Request for Rehearing at 7-8; see also Memorandum: Reminders on evaluating subject matter eligibility of claims under 35 U.S.C. 101 at 4-5 ("[i]n computer-related technologies, examiners can conclude that claims are eligible in Step 2A Prong Two by finding that a claim reflects an improvement to the functioning of a computer or to another technology or technical field, integrating a recited judicial exception into a practical application of the exception"); MPEP § 2106.04(d). Applicant submits that the amended claims meet this standard because the claimed approach is directed towards an improvement to autonomous driving technology. See id. As the present Application makes clear, the claimed approach imparts the technological improvement of enabling a vehicle that is being controlled using a trained machine learning model to adapt to the traffic rules in different geographical locations. See Application at [0150], [0153], [0162], and [0167] - [0168]. As a result, the vehicle can be automatically controlled to drive in a manner that adheres to local traffic rules and is safer than what can typically be achieved using conventional machine learning models. See id. Alternatively, display and/or audio output can be generated directing a user to drive more safely and in accordance with local traffic rules. See id. These improvements to technology are effectuated by the amended claims through the express steps of extracting at least one portion of a set of traffic rules based on the description of the scene and the first plan; generating a first prompt that requests driving instructions and includes the description of the scene, the first plan, and the at least one portion of the set of traffic rules; processing the first prompt via a first trained language model to generate a second plan for driving the vehicle; and performing one or more operations to control the vehicle or generate output via an output device based on the second plan. Thus, the mechanism for effecting the technological improvement is captured explicitly in the amended claims.
Because the amended claims are directed towards an improvement to technology, the amended claims necessarily integrate any purported abstract idea into a practical application and, accordingly, are not abstract. See MPEP §§ 2106.04(d), 2106.05(a)(II).
Because the amended claims do not recite an abstract idea, and because the amended claims recite limitations that integrate any purported abstract idea into a practical application, the amended claims are subject-matter eligible under Step 2A of the 2019 Guidance.
Examiner respectfully disagrees. The claims, given their broadest reasonable interpretation, appear to be directed to the use of a machine learning model within a technological field, rather than a specific improvement to the field itself. In Ex Parte Desjardins, as primarily addressed in Applicant Arguments, above, the panel found, regarding the claims at issue, that:
“When evaluating the claim as a whole, we discern at least the following limitation of independent claim 1 that reflects the improvement: "adjust the first values of the plurality of parameters to optimize performance of the machine learning model on the second machine learning task while protecting performance of the machine learning model on the first machine learning task." We are persuaded that constitutes an improvement to how the machine learning model itself operates, and not, for example, the identified mathematical calculation.”
No such paralleling limitation or similar improvement appears to be present here, in the view of the Examiner. The present claimed invention, as noted above, appears to use the large language model recited as a means to process data and generate a driving plan for the vehicle, rather than effectuate an improvement in a learning model (or similar) itself. Further, regarding Applicant’s assertion that the improvement is of autonomous driving technology, Examiner respectfully asserts that the linkage appears to be one in which a field of use is defined, however the output appears to only be concrete and practical in one of two outcomes. While controlling the vehicle may reflect an improvement in autonomous driving technology, the claim further recites generally “generate output via an output device based on the second plan,” which is not a clear improvement. Under the broadest reasonable interpretation of the term, the claim in the latter encompasses merely processing data and generating an output without integration into substantially more, and therefore does not reflect a clear technological improvement. Thus, Applicant Arguments are not persuasive.
B. Federal Circuit Case Law:
The Federal Circuit has ruled in numerous cases that claims directed towards technological solutions to technological problems are not abstract under the two-step Alice test. Applicant submits that the amended claims are similarly directed towards a technological solution to a technological problem.
In that regard, the present Application makes clear that a technical problem that existed in the prior art prior to the development of the claimed approach was that when a vehicle is controlled by a machine learning model that is trained to adhere to a specific set of traffic rules, the resulting trained model can end up controlling the vehicle in erroneous ways that contravene local traffic rules that are different, resulting in dangerous driving scenarios that raise safety concerns. See Application at [0005]. Accordingly, conventional autonomous vehicles are usually limited to operating primarily in geo-fenced areas where the traffic rules do not change. See id. The present Application also makes clear that the technical advantages of the claimed approach include enabling a vehicle that is being controlled using a trained machine learning model to adapt to the traffic rules in different geographical locations. See Application at [0168]. As a result, the vehicle can be automatically controlled to drive in a manner that adheres to local traffic rules and is safer than what can typically be achieved using conventional machine learning models. See id. at [0150], [0153], [0162], and [0167] - [0168]. Alternatively, display and/or audio output can be generated directing a user to drive more safely and in accordance with local traffic rules. See id. Thus, the claimed approach solves the above technical problem that existed in the prior art.
Examiner respectfully asserts, similarly to the above with respect to the discussion of Ex Parte Desjardins, that the claim merely appears to recite the use of a language model in order to process data and generate an output, rather than reflect a clear improvement to a technical problem. The technical advantages recited above in Applicant Arguments may reflect improvements generally in autonomous driving, however do not appear to reflect improvements in a specific technology as required to satisfy the requirement of a specific technical solution to a technical problem to render the claim patent-eligible under the requisite analysis. The present claims appear to recite the use of technology, specifically the use of a language model, in the application of autonomous driving, however the mere use of the technology to effectuate an outcome does not render the use of the technology an inherent improvement, in view of the outcome of merely generating output via an output device based on the second plan as recited by the present claimed invention. Thus, Applicant Arguments are not persuasive.
Further, each of the amended independent claims recites the limitations of "generating one or more augmented samples based on the one or more samples and one or more augmentation parameters," and "performing one or more operations to train the machine learning model based on the one or more augmented samples and the one or more augmentation parameters to yield a trained machine learning model, wherein the one or more augmentation parameters are input into the machine learning model as conditioning information during at least one operation included in the one or more operations to train the machine learning model." These limitations are specific to imparting the technological improvement of the claimed approach. See id. at [0068] - [0072].
Accordingly, the amended claims are subject-matter eligible under the legal rule set forth in Finjan, Inc. v. Blue Coat Sys., Inc., 879 F.3d 1299 (Fed. Cir. 2018) and McRO, Inc. v. Bandai Namco Games America Inc., 837 F.3d 1299 (Fed. Cir. 2016) (claims that recite specifically limited steps or elements that effect a technological improvement or useful result are not abstract), under the legal rule set forth in Visual Memory LLC v. NVIDIA Corp., 867 F.3d 1253 (Fed. Cir. 2017) (claims directed towards a technological improvement are not abstract), and the legal rule set forth in Weisner v. Google LLC, No. 2021-2228 (Fed. Cir. 2022) and DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245 (Fed. Cir. 2014) (claims directed towards a technical solution to a technical problem necessarily recite more than an abstract idea).
Examiner respectfully disagrees, drawing further reference to Claim 2 of Example 47 of the Subject Matter Eligibility guidelines. In the referenced example claim, the limitations include receiving and discretizing continuous training data, which is then used to train an artificial neural network. Examiner respectfully asserts that said limitations parallel the “generating…” and “performing one or more operations to train…” limitations recited by Applicant above. The mere act of generating data and training a machine learning model based on such does not appear to render the claim patent-eligible, and for the reasons set forth above, the mere use of the machine learning model in the autonomous driving system does not render the claim patent-eligible. Thus, Applicant Arguments with respect to 35 USC 101 are not persuasive.
Regarding 35 USC 103
As amended, claim 1 recites the limitations of generating a first prompt that requests driving instructions and includes the description of the scene, the first plan, and the at least one portion of the set of traffic rules. Claim 1 also recites that the first plan is for driving a vehicle. None of the cited references teaches or suggests these limitations. Therefore, no combination of the cited references can teach or suggest each and every limitation of amended claim 1.
Chandraker discloses a large language model (LLM)-based hybrid planner for autonomous driving. See Chandraker at Abstract. In Chandraker, an LLM is prompted to generate parameters for a rule-based planner based on historical data for vehicles in a road scene. See id. at Abstract, [0004] - [0006]. A trajectory is generated using the parameters, and then a driving action is performed to implement the trajectory. See id.
In the rejections, the Examiner maps the limitations of generating a first prompt that requests driving instructions and includes (1) the description of the scene, (2) the first plan, and (3) the at least one portion of the set of traffic rules, recited in prior claim 1, to the prompt that instructs an LLM to generate achievable subgoals to reach a destination and describes (1) the positions and velocities of other vehicles as well as (2) parameters such as traffic rules, disclosed in Chandraker. See Office Action 10-11. Based on these claim mappings, to teach or suggest the above limitations of amended claim 1, Chandraker would have to disclose that the prompt includes a plan for driving a vehicle. Importantly, Chandraker contains no such teachings or suggestions. Instead, Chandraker discloses the idea of prompting the LLM "with a goal that states the vehicle's current position within a road scene, describes the positions and velocities of other vehicles, and instructs the LLM ... to generate achievable sub-goals ... to reach a target destination." See Chandraker at [0035] - [0037]. Chandraker also discloses that "[t]he prompt may include context that sets parameters, such as rules and formatting for how the output is to be generated." See id. Notably, however, Chandraker is silent with respect to the prompt including a plan for driving a vehicle, as required by the language of amended claim 1. In view of at least these distinctions, Applicant submits that Chandraker cannot be properly interpreted as teaching or suggesting the above limitations of amended claim 1.
Examiner respectfully disagrees. In the present specification, see Paragraphs 0152 & 0153, the “motion plan” is recited to include a future motion trajectory, including waypoints “(e.g., [(x1, y1), … , (xN, yN)]” of the vehicle’s motion. In Chandraker, Paragraphs 0035 – 0036, the LLM is disclosed as being prompted with a goal, including a vehicle’s current position and a target destination. Under the broadest reasonable interpretation of a “motion plan” in the context of the claim and disclosure, Examiner is of the view that the “first plan for driving a vehicle” in the context of the of the claim would be encompassed by at least the definition of a target destination waypoint from the current position of the vehicle that the model of Chandraker is prompted with, effectively defined waypoints of a route for the vehicle to travel between, which may be refined into a more particularized motion plan by the model. Thus, Applicant Arguments with respect to 35 USC 103 are not persuasive.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1 – 3, 5 – 13, & 15 – 20 is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
The determination of whether a claim recites patent ineligible subject matter is a 2 step inquiry.
STEP 1: the claim does not fall within one of the four statutory categories of invention (process, machine, manufacture or composition of matter), see MPEP 2106.03, or
STEP 2: the claim recites a judicial exception, e.g. an abstract idea, without reciting additional elements that amount to significantly more than the judicial exception, as determined using the following analysis: see MPEP 2106.04
STEP 2A (PRONG 1): Does the claim recite an abstract idea, law of nature, or natural phenomenon? see MPEP 2106.04(II)(A)(1)
STEP 2A (PRONG 2): Does the claim recite additional elements that integrate the judicial exception into a practical application? see MPEP 2106.04(II)(A)(2) and 2106.05(a) thru (d) for explanations.
STEP 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception? see MPEP 2106.05
101 Analysis – Step 1
Claim 1 is directed to a method of generating driving instructions for a vehicle (i.e., a process). Therefore, claim 1 is within at least one of the four statutory categories. Similarly, Claims 11 & 20 are respectively directed towards non-transitory computer-readable media and a system for generating driving instructions for a vehicle, respectively, and are also within at least one of the four statutory categories.
101 Analysis – Step 2A, Prong I
Regarding Prong I of the Step 2A analysis, the claims are to be analyzed to determine whether they recite subject matter that falls within one of the follow groups of abstract ideas: a) mathematical concepts, b) certain methods of organizing human activity, and/or c) mental processes. see MPEP 2106(A)(II)(1) and MPEP 2106.04(a)-(c)
Independent claim 20 includes limitations that recite an abstract idea (emphasized below [with the category of abstract idea in brackets]) and will be used as a representative claim for the remainder of the 101 rejection. Claim 20 recites:
A system, comprising:
one or more memories storing instructions; and one or more processors that are coupled to the one or more memories and, when executing the instructions, are configured to:
receive first text that includes a description of a scene and a first plan for driving a vehicle,
extract at least one portion of a set of traffic rules based on the description of the scene and the first plan, [mental process/step]
generate a first prompt that requests driving instructions and includes the description of the scene, the first plan, and the at least one portion of the set of traffic rules, [mental process/step]
process the first prompt via a first trained language model to generate a second plan for driving the vehicle, and [mental process/step]
performing one or more operations to control the vehicle or to generate output via an output device based on the second plan. [mental process/step]
The examiner submits that the foregoing bolded limitation(s) constitute a “mental process” because under its broadest reasonable interpretation, the claim covers performance of the limitation in the human mind. For example, “extract…” in the context of this claim encompasses a person looking at data collected regarding a scene and first plan, and forming a simple judgement as to the traffic rules. Similarly, “generate a first prompt…” in the context of the claim encompasses a person generating a prompt to request driving instructions based on data collected, which is a mental process under its broadest reasonable interpretation. Further, “process…” in the context of the claim encompasses utilizing a trained language model to generate a plan for driving a vehicle based on an input prompt determined in the previous step, which is a mental process of evaluating data and forming a simple judgment. Finally, “based on the second plan…” in the context of the claim encompasses a person evaluating data collected regarding the second plan, and determining an output or vehicle control operations based on such, which is a process of evaluating data collected and forming a simple judgment as to the driving instructions required to execute the plan, which is a mental process under its broadest reasonable interpretation. Accordingly, the claim recites at least one abstract idea.
101 Analysis – Step 2A, Prong II
Regarding Prong II of the Step 2A analysis, the claims are to be analyzed to determine whether the claim, as a whole, integrates the abstract into a practical application. see MPEP 2106.04(II)(A)(2) and MPEP 2106.04(d)(2). It must be determined whether any additional elements in the claim beyond the abstract idea integrate the exception into a practical application in a manner that imposes a meaningful limit on the judicial exception. The courts have indicated that additional elements merely using a computer to implement an abstract idea, adding insignificant extra solution activity, or generally linking use of a judicial exception to a particular technological environment or field of use do not integrate a judicial exception into a “practical application.”
In the present case, the additional limitations beyond the above-noted abstract idea are as follows (where the underlined portions are the “additional limitations” [with a description of the additional limitations in brackets], while the bolded portions continue to represent the “abstract idea”.):
A system, comprising: [Apply it, 2106.05(f)]
one or more memories storing instructions; and one or more processors that are coupled to the one or more memories and, when executing the instructions, are configured to: [applying the abstract idea using generic computing module, Apply it 2106.05(f)]
receive first text that includes a description of a scene and a first plan for driving a vehicle, pre-solution activity (data gathering), 2106.05(g)]
extract at least one portion of a set of traffic rules based on the description of the scene and the first plan,
generate a first prompt that requests driving instructions and includes the description of the scene, the first plan, and the at least one portion of the set of traffic rules,
process the first prompt via a first trained language model to generate a second plan for driving the vehicle, and
performing one or more operations to control the vehicle or to generate output via an output device [See notes below, insignificant post-solution activity (displaying results of the mental process) 2106.05(g)] based on the second plan.
For the following reason(s), the examiner submits that the above identified additional limitations do not integrate the above-noted abstract idea into a practical application.
Regarding the additional limitations of “one or more memories…,” “one or more processors…,” and “receive first text…,” the examiner submits that these limitations are insignificant extra-solution activities that merely use a computer to perform the process. In particular, the “receive first text…” step is recited at a high level of generality (i.e. as a general means of gathering a description of a scene and a first plan for driving a vehicle), and amounts to mere data gathering, which is a form of insignificant extra-solution activity. Further, the “one or more memories…” and “one or more processors…” are recited at a high-level of generality (i.e., as generic processor(s) and memory for performing a computer function) such that it amounts no more than mere instructions to apply the exception using a generic computer component.
Finally, “performing one or more operations to control the vehicle or to generate output via an output device” in the context of the claim encompasses two options as to the results of the aforementioned steps. While “performing one or more operations to control the vehicle” is a practical application of the use of the language model, and would thus be patent-eligible, the limitation “generate output via an output device” encompasses the generic output of data, which is insignificant extra-solution activity that does not render the claim patent-eligible. As either the one or more operations or the output may be utilized independently to fulfill the requirements of the present claimed invention, the claim under its broadest reasonable interpretation is not patent-eligible, as the possible prong of generically outputting information via an output device does not recite substantially more than the abstract idea.
Thus, taken alone, the additional elements do not integrate the abstract idea into a practical application. Further, looking at the additional limitation(s) as an ordered combination or as a whole, the limitation(s) add nothing that is not already present when looking at the elements taken individually. For instance, there is no indication that the additional elements, when considered as a whole, reflect an improvement in the functioning of a computer or an improvement to another technology or technical field, apply or use the above-noted judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition, implement/use the above-noted judicial exception with a particular machine or manufacture that is integral to the claim, effect a transformation or reduction of a particular article to a different state or thing, or apply or use the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is not more than a drafting effort designed to monopolize the exception. see MPEP § 2106.05. Accordingly, the additional limitation(s) do/does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
101 Analysis – Step 2B
Regarding Step 2B of the Revised Guidance, representative independent claim 1 does not include additional elements (considered both individually and as an ordered combination) that are sufficient to amount to significantly more than the judicial exception for the same reasons to those discussed above with respect to determining that the claim does not integrate the abstract idea into a practical application. As discussed above with respect to integration of the abstract idea into a practical application, the additional element of using one or more memories and processors to perform the steps of the mental process amounts to nothing more than mere instructions to apply the exception using a generic computer component. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. And as discussed above, the additional limitations of “receive first text…” and “performing one or more operations to control the vehicle or to generate output via an output device,” the examiner submits that these limitations are insignificant extra-solution activity.
Dependent claim(s) 2 – 3, 5 – 10, 12 – 13, & 15 – 19 do not recite any further limitations that cause the claim(s) to be patent eligible. Rather, the limitations of dependent claims are directed toward additional aspects of the judicial exception and do not integrate the judicial exception into a practical application. Specifically:
Claim 2 recites wherein a second text is received indicating a situation which the prompt includes, and the set of traffic rules being based on said situation, which is the receipt of data and the evaluation of which under its broadest reasonable interpretation.
Claim 3 recites wherein the extracting of traffic rules comprises generating a prompt to extract keywords, which is a mental process of evaluating data under its broadest reasonable interpretation.
Claim 5 recites wherein the driving instructions are output using a speaker device, which is the insignificant extra-solution activity of outputting the results of the mental process using generic components that are well known in the art under the broadest reasonable interpretation of the claim.
Claim 6 recites wherein the driving instructions are output using a display device, which is the insignificant extra-solution activity of outputting the results of the mental process using generic components that are well known in the art under the broadest reasonable interpretation of the claim.
Claim 7 recites wherein a vision language model is utilized to process image data to generate a description of the scene, which is the receipt of data and the evaluation of such under its broadest reasonable interpretation.
Claim 8 recites wherein a first plan is generated based on a second trained language model and second text indicating sensor data/status information associated with the vehicle, which, similarly to as set forth in the independent claim(s), is the receipt of data and the evaluation of which under its broadest reasonable interpretation.
Claim 9 recites wherein the set of traffic rules are included in a driving handbook, which merely recites the insignificant extra-solution activity of receiving data under the broadest reasonable interpretation of the claim.
Claim 10 recites wherein the first trained language model comprises a trained large language model, which merely recites a specific mathematical evaluation framework with which data is evaluated, which is an abstract idea under the broadest reasonable interpretation of the claim.
Claim 19 recites wherein the first plan and text are received from either a user or trained machine learning model, which merely recites the insignificant extra-solution activity of receiving data under the broadest reasonable interpretation of the claim.
Claims 12, 13, & 15 – 18 recite substantially similar limitations as those found in Claims 2, 3, & 5 – 8, and are rejected under similar rationale as set forth above.
Therefore, dependent claims 2 – 3, 5 – 10, 12 – 13, & 15 – 19 are not patent eligible under the same rationale as provided for in the rejection of Independent Claims 1, 11, & 20.
Therefore, claim(s) 1 – 3, 5 – 13, & 15 – 20 is/are ineligible under 35 USC §101.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1, 4, 9 - 11, 14, & 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chandraker (US 2025/0145176 A1) in view of Wang (US 2022/0289238 A1).
Regarding Claim 1:
Chandraker discloses: A computer-implemented method for controlling a vehicle, the method comprising: (Chandraker discloses in at least Paragraphs 0004 & 0005 a method for operating a vehicle executable by a hardware processor and memory, the method including the prompting of a large language model to generate parameters for a rule-based planner, and the generation of a vehicle trajectory using the planner)
receiving first text that includes a description of a scene and… (Chandraker discloses in at least Paragraphs 0027, 0032, & 0036 wherein text-based scene descriptions, including vehicular positioning, traffic light status, and lane information, as well as a goal that states the target destination of the vehicle may be provided as an input to the LLM to generate a navigational plan for the vehicle [i.e. receiving first text that includes a description of a scene])
generating a first prompt that requests driving instructions and includes the description of the scene, the first plan, and the at least one portion of the set of traffic rules; (Chandraker discloses in at least Paragraph 0036 wherein a prompt may be supplied to the LLM from an LLM reasoning module, the prompt instructing the LLM to generate achievable sub-goals to reach a destination [i.e. generating a first prompt that requests driving instructions]. At least Paragraphs 0035 & 0036 of Chandraker further teach wherein the prompt may describe the positions and velocities of other vehicles [i.e. the prompt includes the description of the scene], include parameters, such as traffic rules [i.e. the prompt includes at least one portion of the set of traffic rules], and provide the target destination of the vehicle [i.e. a first plan])
processing the first prompt via a first trained language model to generate a second plan for driving the vehicle; and (Chandraker discloses in at least Paragraphs 0041 – 0043 wherein at each time step a prompt is issued from the LLM interface to the LLM, the prompt containing up-to-date scene information and position history for surrounding objects, as well as the prompt information set forth above. The LLM then proposes a trajectory for the vehicle based on the issued prompt, which may be scored and selected if the scoring meets a predefined criteria [i.e. a second plan for driving the vehicle is generated based on processing the first prompt via a first trained language model]. At least Paragraphs 0060 – 0062 of Chandraker further disclose the generation of trajectories by the rule-based planner based on the parameters output from the LLM when said LLM is prompted)
performing one or more operations to control the vehicle or to generate output via an output device based on the second plan. (Chandraker discloses in at least Paragraphs 0006 & 0060 – 0062 wherein based on the selected trajectory [i.e. based on the second plan], an action may be taken to execute the selected trajectory, including triggering the ECU to perform a driving action to implement the trajectory, including through braking, steering, and/or acceleration actions [i.e. performing one or more operations to control the vehicle based on the second plan])
Chandraker however appears to be silent regarding:
receiving… a first plan for driving a vehicle;
extracting at least one portion of a set of traffic rules based on the description of the scene and the first plan;
However Wang teaches wherein traffic rules may be extracted based on associating keywords tagged to rules stored in a database with vehicle state and environmental information to retrieve relevant traffic rules to the current context.
receiving… a first plan for driving a vehicle; (However Wang teaches in at least Paragraphs 0033 & 0057 wherein an external device may receive information, including information on environmental conditions and vehicle states, which may include traveling direction, speed, acceleration, currently travelling lane and motion trajectory [i.e. receiving a first plan for driving a vehicle] as taught in at least Paragraph 0038 of Wang)
extracting at least one portion of a set of traffic rules based on the description of the scene and the first plan; (However Wang teaches in at least Paragraphs 0046 – 0050 wherein a database of traffic laws may be pre-processed, including by adding tags or labels to the information corresponding to category-based keywords to traffic rules. Wang further teaches in at least Paragraph 0057 wherein based on the vehicle state, which may include traveling direction, speed, acceleration, currently travelling lane and motion trajectory as taught in at least Paragraph 0038 [i.e. the first plan] and environmental information [i.e. the description of the scene], an information category is obtained and utilized to retrieve one or more traffic rules related to the information category the acquired information belongs to from a traffic rule database [i.e. extracting at least one portion of a set of traffic rules based on the description of the scene and the first plan)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the extraction of traffic rules based on the determined vehicle state and environmental information as taught by Wang.
The motivation to do so is that, as acknowledged by Wang in at least Paragraph 0050, by tagging the traffic rules in the database with keywords associated with the situation, identifying the situation based on the vehicle state, and evaluating the present state & environment of the vehicle to compare to the identified keywords, the efficiency of retrieving relevant traffic rules may be improved.
Regarding Claim 4:
The computer-implemented method of claim 1, wherein performing one or more operations control the vehicle comprises: generating first driving instructions based on the second plan; generating, by a trained planning model and based on the first driving instructions, one or more trajectories for the vehicle; and causing one or more operations to control the vehicle to be performed based on the one or more trajectories.
Chandraker discloses in at least Paragraphs 0006 & 0060 – 0062 wherein the rule based planner [i.e. a trained planning model] may generate a set of trajectories [i.e. generating first driving instructions based on the second plan, generating, by the trained planning model and based on the first driving instructions, one or more trajectories for the vehicle], select a trajectory, and an action may be taken to execute the selected trajectory, including triggering the ECU to perform a driving action to implement the trajectory [i.e. causing one or more operations to control the vehicle to be performed based on the one or more trajectories].
Regarding Claim 9:
The computer-implemented method of claim 1, wherein the set of traffic rules are included in a driving handbook.
Chandraker appears to be silent regarding wherein the set of traffic rules are included in a driving handbook.
However Wang teaches in at least Paragraphs 0041 & 0042 wherein the traffic rule information obtained from the external device may include traffic rules documented in a traffic rule manual or guide [i.e. the set of traffic rules are included in a driving handbook]
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the acquisition of traffic rules from a manual or handbook as taught by Wang.
The motivation to do so is that, as acknowledged by Wang in at least Paragraphs 0041 & 0042, the system may acquire traffic rules that are not expected to change over time through official documentation, improving the acquisition of relevant traffic rules.
Regarding Claim 10:
The computer-implemented method of claim 1, wherein the first trained language model comprises a trained large language model.
Chandraker discloses in at least Paragraphs 0018, 0029, & 0030 wherein a large language model [LLM] may be used to determine behavioral sub-goals [i.e. wherein the first trained language model comprises a trained large language model].
Regarding Claim 11:
Chandraker discloses: One or more non-transitory computer-readable media storing instructions that, when executed by at least one processor, cause the at least one processor to perform the steps of: (Chandraker discloses in at least Paragraphs 0005 & 0006 a system that may include a hardware processor and memory that stores a computer program configured to be executed by the hardware processor [i.e. one or more non-transitory computer-readable media storing instructions executable by the processor] to prompt the LLM disclosed for operating a vehicle, including prompting the LLM to generate parameters for a rule-based planner based on historical data for vehicles in a road scene, generating a trajectory using the parameters, and performing a driving action to implement the trajectory)
receiving first text that includes a description of a scene and… (Chandraker discloses in at least Paragraphs 0027, 0032, & 0036 wherein text-based scene descriptions, including vehicular positioning, traffic light status, and lane information, as well as a goal that states the target destination of the vehicle may be provided as an input to the LLM to generate a navigational plan for the vehicle [i.e. receiving first text that includes a description of a scene])
generating a first prompt that requests driving instructions and includes the description of the scene, the first plan, and the at least one portion of the set of traffic rules; (Chandraker discloses in at least Paragraph 0036 wherein a prompt may be supplied to the LLM from an LLM reasoning module, the prompt instructing the LLM to generate achievable sub-goals to reach a destination [i.e. generating a first prompt that requests driving instructions]. At least Paragraphs 0035 & 0036 of Chandraker further teach wherein the prompt may describe the positions and velocities of other vehicles [i.e. the prompt includes the description of the scene], include parameters, such as traffic rules [i.e. the prompt includes at least one portion of the set of traffic rules], and provide the target destination of the vehicle [i.e. the first plan])
processing the first prompt via a first trained language model to generate a second plan for driving the vehicle; and (Chandraker discloses in at least Paragraphs 0041 – 0043 wherein at each time step a prompt is issued from the LLM interface to the LLM, the prompt containing up-to-date scene information and position history for surrounding objects, as well as the prompt information set forth above. The LLM then proposes a trajectory for the vehicle based on the issued prompt, which may be scored and selected if the scoring meets a predefined criteria [i.e. a second plan for driving the vehicle is generated based on processing the first prompt via a first trained language model]. At least Paragraphs 0060 – 0062 of Chandraker further disclose the generation of trajectories by the rule based planner based on the parameters output from the LLM when said LLM is prompted)
performing one or more operations to control the vehicle or to display output via a display device based on the second plan. (Chandraker discloses in at least Paragraphs 0006 & 0060 – 0062 wherein based on the selected trajectory [i.e. based on the second plan], an action may be taken to execute the selected trajectory, including triggering the ECU to perform a driving action to implement the trajectory, including through braking, steering, and/or acceleration actions [i.e. performing one or more operations to control the vehicle based on the second plan])
Chandraker however appears to be silent regarding:
receiving… a first plan for driving a vehicle;
extracting at least one portion of a set of traffic rules based on the description of the scene and the first plan;
However Wang teaches wherein traffic rules may be extracted based on associating keywords tagged to rules stored in a database with vehicle state and environmental information to retrieve relevant traffic rules to the current context.
receiving… a first plan for driving a vehicle; (However Wang teaches in at least Paragraphs 0033 & 0057 wherein an external device may receive information, including information on environmental conditions and vehicle states, which may include traveling direction, speed, acceleration, currently travelling lane and motion trajectory as taught in at least Paragraph 0038 of Wang [i.e. receiving a first plan for driving a vehicle])
extracting at least one portion of a set of traffic rules based on the description of the scene and the first plan; (However Wang teaches in at least Paragraphs 0046 – 0050 wherein a database of traffic laws may be pre-processed, including by adding tags or labels to the information corresponding to category-based keywords to traffic rules. Wang further teaches in at least Paragraph 0057 wherein based on the vehicle state, which may include traveling direction, speed, acceleration, currently travelling lane and motion trajectory as taught in at least Paragraph 0038 [i.e. the first plan] and environmental information [i.e. the description of the scene], an information category is obtained and utilized to retrieve one or more traffic rules related to the information category the acquired information belongs to from a traffic rule database [i.e. extracting at least one portion of a set of traffic rules based on the description of the scene and the first plan)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the extraction of traffic rules based on the determined vehicle state and environmental information as taught by Wang.
The motivation to do so is that, as acknowledged by Wang in at least Paragraph 0050, by tagging the traffic rules in the database with keywords associated with the situation, identifying the situation based on the vehicle state, and evaluating the present state & environment of the vehicle to compare to the identified keywords, the efficiency of retrieving relevant traffic rules may be improved.
Regarding Claim 14:
Claim 14 recites substantially similar limitations as those found in Claim 4, above, and is rejected under similar rationale.
Regarding Claim 20:
Chandraker discloses: A system, comprising: (Chandraker discloses in at least Paragraph 0005 a system for operating a vehicle, including prompting an LLM to generate parameters for a rule-based planner based on historical data for vehicles in a road scene, generating a trajectory using the parameters, and performing a driving action to implement the trajectory)
one or more memories storing instructions; and one or more processors that are coupled to the one or more memories and, when executing the instructions, are configured to: (Chandraker discloses in at least Paragraphs 0005 & 0006 wherein the system for operating the vehicle may include a hardware processor and memory that stores a computer program configured to be executed by the hardware processor to prompt the LLM disclosed)
receive first text that includes a description of a scene and… (Chandraker discloses in at least Paragraphs 0027, 0032, & 0036 wherein text-based scene descriptions, including vehicular positioning, traffic light status, and lane information, as well as a goal that states the target destination of the vehicle may be provided as an input to the LLM to generate a navigational plan for the vehicle [i.e. receiving first text that includes a description of a scene])
generate a first prompt that requests driving instructions and includes the description of the scene, the first plan, and the at least one portion of the set of traffic rules, (Chandraker discloses in at least Paragraph 0036 wherein a prompt may be supplied to the LLM from an LLM reasoning module, the prompt instructing the LLM to generate achievable sub-goals to reach a destination [i.e. generating a first prompt that requests driving instructions]. At least Paragraphs 0035 & 0036 of Chandraker further teach wherein the prompt may describe the positions and velocities of other vehicles [i.e. the prompt includes the description of the scene], include parameters, such as traffic rules [i.e. the prompt includes at least one portion of the set of traffic rules], and provide the target destination of the vehicle [i.e. the first plan])
process the first prompt via a first trained language model to generate a second plan for driving the vehicle, and (Chandraker discloses in at least Paragraphs 0041 – 0043 wherein at each time step a prompt is issued from the LLM interface to the LLM, the prompt containing up-to-date scene information and position history for surrounding objects, as well as the prompt information set forth above. The LLM then proposes a trajectory for the vehicle based on the issued prompt, which may be scored and selected if the scoring meets a predefined criteria [i.e. a second plan for driving the vehicle is generated based on processing the first prompt via a first trained language model]. At least Paragraphs 0060 – 0062 of Chandraker further disclose the generation of trajectories by the rule based planner based on the parameters output from the LLM when said LLM is prompted)
perform one or more operations to control the vehicle or to display output via a display device based on the second plan. (Chandraker discloses in at least Paragraphs 0006 & 0060 – 0062 wherein based on the selected trajectory [i.e. based on the second plan], an action may be taken to execute the selected trajectory, including triggering the ECU to perform a driving action to implement the trajectory, including through braking, steering, and/or acceleration actions [i.e. performing one or more operations to control the vehicle based on the second plan])
Chandraker however appears to be silent regarding:
receive… a first plan for driving a vehicle,
extract at least one portion of a set of traffic rules based on the description of the scene and the first plan,
However Wang teaches wherein traffic rules may be extracted based on associating keywords tagged to rules stored in a database with vehicle state and environmental information to retrieve relevant traffic rules to the current context.
receive… a first plan for driving a vehicle, (However Wang teaches in at least Paragraphs 0033 & 0057 wherein an external device may receive information, including information on environmental conditions and vehicle states, which may include traveling direction, speed, acceleration, currently travelling lane and motion trajectory as taught in at least Paragraph 0038 of Wang [i.e. receiving a first plan for driving a vehicle])
extract at least one portion of a set of traffic rules based on the description of the scene and the first plan, (However Wang teaches in at least Paragraphs 0046 – 0050 wherein a database of traffic laws may be pre-processed, including by adding tags or labels to the information corresponding to category-based keywords to traffic rules. Wang further teaches in at least Paragraph 0057 wherein based on the vehicle state, which may include traveling direction, speed, acceleration, currently travelling lane and motion trajectory as taught in at least Paragraph 0038 [i.e. the first plan] and environmental information [i.e. the description of the scene], an information category is obtained and utilized to retrieve one or more traffic rules related to the information category the acquired information belongs to from a traffic rule database [i.e. extracting at least one portion of a set of traffic rules based on the description of the scene and the first plan)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the extraction of traffic rules based on the determined vehicle state and environmental information as taught by Wang.
The motivation to do so is that, as acknowledged by Wang in at least Paragraph 0050, by tagging the traffic rules in the database with keywords associated with the situation, identifying the situation based on the vehicle state, and evaluating the present state & environment of the vehicle to compare to the identified keywords, the efficiency of retrieving relevant traffic rules may be improved.
Claim(s) 2, 3, 8, 12, 13, & 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chandraker (US 2025/0145176 A1) in view of Wang (US 2022/0289238 A1) as applied to claims 1 & 11 above, and further in view of Jang (US 2025/0145174 A1).
Regarding Claim 2:
The computer-implemented method of claim 1, further comprising receiving second text that indicates a situation, wherein extracting the at least one portion of the set of traffic rules is further based on the situation, and wherein the prompt further includes the situation.
Chandraker does not appear to specifically disclose receiving second text that indicates a situation wherein extracting the at least one portion of the set of traffic rules is further based on the situation.
However Jang teaches in at least Paragraphs 0050, 0056, & 0059 wherein a driving scenario [i.e. situation] data may be obtained and used to determine traffic law(s) associated with the situation, such as receiving a scenario in which a vehicle turns right, and the relevant law(s) may be determined for said right turn scenario, such as stopping requirements [i.e. extracting the at least one portion of the set of traffic rules is further based on the situation]. Jang however also appears to be silent regarding wherein the situation is received via a second text, as well as wherein the prompt includes the situation.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the determination of traffic laws associated with a specific situation input as taught by Jang.
The motivation to do so is that, as acknowledged by Jang in at least Paragraph 0082, the vehicle may drive in compliance with road traffic laws for each region and determined situation, improving the operation and safety of the vehicle.
However Wang teaches in at least Paragraphs 0046 – 0050 wherein a database of traffic laws may be pre-processed, including by adding tags or labels to the information corresponding to category-based keywords to traffic rules. Wang further teaches in at least Paragraph 0057 wherein based on the vehicle state and environmental information, an information category is obtained and utilized to retrieve one or more traffic rules related to the information category the acquired information belongs to from a traffic rule database [i.e. receiving second text that indicates a situation… and wherein the prompt further includes the situation].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the tagging of traffic rule data with keywords for retrieval as taught by Wang.
The motivation to do so is that, as acknowledged by Wang in at least Paragraph 0050, by tagging the traffic rules in the database with keywords associated with the situation, the efficiency of retrieving relevant traffic rules may be improved.
Regarding Claim 3:
The computer-implemented method of claim 1, wherein extracting the at least one portion of the set of a traffic rules comprises: generating a second prompt that asks for traffic phrases and includes the description of the scene and the first plan; processing the second prompt via the first trained language model to extract one or more keywords from the description of the scene and the first plan; and extracting, from the set of traffic rules, one or more paragraphs that include at least a first keyword that is included in the one or more keywords.
Chandraker does not appear to specifically disclose wherein a portion of the set of traffic rules are extracted based on prompting or keywords.
However Jang teaches in at least Paragraphs 0050, 0056, & 0059 wherein a driving scenario [i.e. situation] data may be obtained and used to determine traffic law(s) associated with the situation, such as receiving a scenario in which a vehicle turns right, and the relevant law(s) may be determined for said right turn scenario, such as stopping requirements [i.e. extracting the at least one portion of the set of traffic rules is further based on the situation]. Jang further teaches in at least Paragraphs 0006, 0025, & 0048 – 0050 wherein traffic laws may be understood by inputting [i.e. prompting] a LLM with a dataset relating to driving from a database, from which inferences may be made regarding the driving rules [i.e. generating a second prompt that asks for traffic phrases and includes the description of the scene and the first plan; processing the second prompt via the first trained language model]. Jang however also appears to be silent regarding wherein the situation is received via a second text, as well as wherein the prompt includes the situation.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the extraction of traffic rules through the use of an LLM as taught by Jang.
The motivation to do so is that, as acknowledged by Jang in at least Paragraphs 0051 & 0062, traffic laws may be dynamically updated through the use of a learning model, improving the updating and learning of traffic laws by the vehicle based on the current environment.
However Wang teaches in at least Paragraphs 0046 – 0050 wherein a database of traffic laws may be pre-processed, including by adding tags or labels to the information corresponding to category-based keywords to traffic rules. Wang further teaches in at least Paragraph 0057 wherein based on the vehicle state and environmental information, an information category is obtained [i.e. one or more keywords is extracted from the description of the scene and the first plan] and utilized to retrieve one or more traffic rules related to the information category the acquired information belongs to from a traffic rule database [i.e. extract one or more keywords from the description of the scene and the first plan; and extracting, from the set of traffic rules, one or more paragraphs that include at least a first keyword that is included in the one or more keywords].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the tagging of traffic rule data with keywords for retrieval as taught by Wang.
The motivation to do so is that, as acknowledged by Wang in at least Paragraph 0050, by tagging the traffic rules in the database with keywords associated with the situation, the efficiency of retrieving relevant traffic rules may be improved.
Regarding Claim 8:
The computer-implemented method of claim 1, further comprising generating, via a second trained language model, the first plan based on second text that indicates at least one of sensor data or status information associated with the vehicle.
Chandraker appears to be silent regarding wherein the first plan is generated via a second trained language model based on second text that indicates at least one of sensor data or status information associated with the vehicle.
However Jang teaches in at least Paragraphs 0052 & 0053 and Figure 2, below, wherein a first LLM may receive various natural language inputs, understand the meaning, intention, and context of input sentences, [i.e. second text that indicates at least one of sensor data or status information associated with the vehicle] and output a proper corresponding result to the behavior/trajectory planner as taught in at least Paragraph 0056 of Jang [i.e. the first plan is generated via a second trained language model], the tuning of the LLM to learn traffic laws being made based on the output of the behavior-and-trajectory planner from the first LLM as taught in at least Paragraph 0048 of Jang.
PNG
media_image1.png
456
680
media_image1.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the determination of the first plan via a LLM as taught by Jang.
The motivation to do so is that, as acknowledged by Jang in at least Paragraphs 0053 – 0055, the behavior and trajectory planner may be updated based on an initial plan output, as well as the scene information, improving the learning and implementation of traffic rules and optimal behaviors by the vehicle.
Regarding Claim 12:
Claim 12 recites substantially similar limitations as those found in Claim 2, above, and is rejected under similar rationale.
Regarding Claim 13:
Claim 13 recites substantially similar limitations as those found in Claim 3, above, and is rejected under similar rationale.
Regarding Claim 18:
Claim 18 recites substantially similar limitations as those found in Claim 8, above, and is rejected under similar rationale.
Claim(s) 5, 6, 15, & 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chandraker (US 2025/0145176 A1) in view of Wang (US 2022/0289238 A1) as applied to claims 1 & 11 above, and further in view of Palanisamy (US 2023/0375351 A1).
Regarding Claim 5:
The computer-implemented method of claim 1, wherein performing one or more operations to generate the output comprises: generating first driving instructions based on the second plan; transmitting the first driving instructions to a driver via a speaker device.
Chandraker discloses in at least Paragraphs 0006 & 0060 – 0062 wherein based on the selected trajectory [i.e. based on the second plan], an action may be taken to execute the selected trajectory, including triggering the ECU to perform a driving action to implement the trajectory, including through braking, steering, and/or acceleration actions [i.e. generating first driving instructions based on the second plan]. Chandraker however does not appear to specifically disclose wherein the driving instructions are transmitted to a driver via a speaker device.
However Palanisamy teaches in at least Paragraphs 0017 & 0038 wherein the vehicle may include a speaker that provides auditory turn-by-turn instructions for a user to follow, the instructions being determined by way of a machine learning model, which may include a language model, as taught in at least Paragraphs 0092 & 0095 of Palanisamy.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the provision of turn-by-turn driving instructions via a speaker as taught by Palanisamy.
The motivation to do so is that, as acknowledged by Palanisamy in at least Paragraph 0038, the driving instructions may be presented to a user for driving and verification, improving the communication of determined driving instructions to a driver of a vehicle for execution.
Regarding Claim 6:
The computer-implemented method of claim 1, wherein performing one or more operations to generate the output comprises: generating first driving instructions based on the second plan; and displaying the first driving instructions to a driver via a display device.
Chandraker discloses in at least Paragraphs 0006 & 0060 – 0062 wherein based on the selected trajectory [i.e. based on the second plan], an action may be taken to execute the selected trajectory, including triggering the ECU to perform a driving action to implement the trajectory, including through braking, steering, and/or acceleration actions [i.e. generating first driving instructions based on the second plan]. Chandraker however does not appear to specifically disclose wherein the driving instructions are transmitted to a driver via a display device.
However Palanisamy teaches in at least Paragraphs 0017 & 0038 wherein the vehicle may include a digital display that presents visual information such as an overview map or turn-by-turn instructions for a user to follow, the instructions being determined by way of a machine learning model, which may include a language model, as taught in at least Paragraphs 0092 & 0095 of Palanisamy.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the provision of turn-by-turn driving instructions via a visual display as taught by Palanisamy.
The motivation to do so is that, as acknowledged by Palanisamy in at least Paragraph 0038, the driving instructions may be presented to a user for driving and verification, improving the communication of determined driving instructions to a driver of a vehicle for execution.
Regarding Claim 15:
Claim 15 recites substantially similar limitations as those found in Claim 5, above, and is rejected under similar rationale.
Regarding Claim 16:
Claim 16 recites substantially similar limitations as those found in Claim 6, above, and is rejected under similar rationale.
Claim(s) 7 & 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chandraker (US 2025/0145176 A1) in view of Wang (US 2022/0289238 A1) as applied to claims 1 & 11 above, and further in view of Hausman (US 2023/0311335 A1).
Regarding Claim 7:
The computer-implemented method of claim 1, further comprising processing image data associated with the vehicle using a vision language model to generate the description of the scene.
Chandraker appears to be silent regarding wherein the description of the scene is generated by processing image data using a vision language model.
However Hausman teaches in at least Paragraphs 0062 & 0095 wherein an image obtained by a robot may be processed by a trained model to generate an image embedding, including scene descriptors describing objects detected in the environment, such as “keys,” “human,” “table,” and the like, which are incorporated into a LLM prompt used to generate a desired output for a robot as taught in at least Paragraphs 0062 & 0063 of Hausman.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the use of a trained model to generate embeddings of objects describing a scene as taught by Hausman.
The motivation to do so is that, as acknowledged by Hausman in at least Paragraphs 0062 & 0063, objects may be determined and categorized in an environment, improving the explanation of the scene to the LLM prompt that said prompt uses as a basis to generate an output.
Regarding Claim 17:
Claim 17 recites substantially similar limitations as those found in Claim 7, above, and is rejected under similar rationale.
Claim(s) 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chandraker (US 2025/0145176 A1) in view of Wang (US 2022/0289238 A1) as applied to claim 11 above, and further in view of Jang (US 2025/0145174 A1) and Hausman (US 2023/0311335 A1).
Regarding Claim 19:
The one or more non-transitory computer-readable media of claim 11, wherein the first text and the first plan are received from at least one of a user or one or more trained machine learning models.
Chandraker does not appear to specifically disclose wherein the first text and plan are received from at least one of a user or one or more trained machine learning models.
However Jang teaches in at least Paragraphs 0052 & 0053 and Figure 2, below, wherein a first LLM may receive various natural language inputs, understand the meaning, intention, and context of input sentences, output a proper corresponding result to the behavior/trajectory planner as taught in at least Paragraph 0056 of Jang [i.e. the first plan is received from one or more trained machine learning models], the tuning of the LLM to learn traffic laws being made based on the output of the behavior-and-trajectory planner from the first LLM as taught in at least Paragraph 0048 of Jang.
PNG
media_image1.png
456
680
media_image1.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the determination of the first plan via a LLM as taught by Jang.
The motivation to do so is that, as acknowledged by Jang in at least Paragraphs 0053 – 0055, the behavior and trajectory planner may be updated based on an initial plan output, as well as the scene information, improving the learning and implementation of traffic rules and optimal behaviors by the vehicle.
However Hausman teaches in at least Paragraphs 0062 & 0095 wherein an image obtained by a robot may be processed by a trained model to generate an image embedding, including scene descriptors describing objects detected in the environment, such as “keys,” “human,” “table,” and the like, which are incorporated into a LLM prompt used to generate a desired output for a robot as taught in at least Paragraphs 0062 & 0063 of Hausman [i.e. the first text is received from one or more trained machine learning models].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Chandraker by incorporating the use of a trained model to generate embeddings of objects describing a scene as taught by Hausman.
The motivation to do so is that, as acknowledged by Hausman in at least Paragraphs 0062 & 0063, objects may be determined and categorized in an environment, improving the explanation of the scene to the LLM prompt that said prompt uses as a basis to generate an output.
Conclusion
The following prior art made of record but not relied upon is considered pertinent to the Applicant’s disclosure:
Neindorf (US 2022/0048535 A1): Neindorf recites a path planning system for a vehicle including the receipt of environmental data associated with a detected environment, generating goal states, and generating candidate trajectories for the vehicle based on the goal state. Various machine learning models may be utilized to determine and select candidate trajectories.
Xin (CN117734728A): Xin recites a autonomous driving system for a vehicle, including the use of a LLM and prompting to determine driving instructions for the vehicle. The driving preferences of an operator may be taken into account, along with driving scene information, and the LLM may output driving decisions for the vehicle.
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CHRISTOPHER RYAN CARDIMINO whose telephone number is (571)272-2759. The examiner can normally be reached M-Th 8:30-5:00.
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, Ramya Burgess can be reached at (571)272-6011. 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.
/CHRISTOPHER R CARDIMINO/Examiner, Art Unit 3661
/RAMYA P BURGESS/Supervisory Patent Examiner, Art Unit 3661
