DETAILED ACTION
This action is in response the communications filed on 05/26/2026 in which claims 1, 6-7, 9, 14-15, 17-25 and 30 are amended and claims 1-30 are pending.
--
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 .
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.
Claims 1-30 are rejected under 35 U.S.C. 112(b) or pre-AIA 35 U.S.C. 112, 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.
Claims 1, 9, 17 and 25 recite the limitation “generating… a belief of the environment based on the observation of the environment….” There is insufficient antecedent basis for the limitation “the observation of the environment” in the claim. For examination purposes examiner has interpreted “the observation of the environment” to be “an observation of the environment.”
Claims 2-8, 10-16, 18-24 and 26-30 are also rejected due to their dependency on a rejected claim.
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 1-30 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more
Step 1: Claims 1-8 recite a method. Claims 9-16 recite an apparatus. Claims 17-24 recites a device. Claims 25-30 recite a non-transitory medium. Therefore, claims 1-8 are directed to a process, claims 9-16 and 17-24 are directed to a machine, and claims 25-30 are directed to a manufacture.
With respect to claims 1, 9, 17 and 25:
2A Prong 1: The claim recites a judicial exception.
generating… a belief of the environment based on the observation of the environment and data associated with prior actions of the robotic device in the environment, the belief representing information of the environment not directly observable by the one or more sensors; and (mental process – evaluation or judgement,--- generating a belief based the observation and data)
2A Prong 2: The judicial exception is not integrated into a practical application.
(claim 17) one or more processors; and one or more memories coupled with the one or more processors and storing instructions operable, when executed by the one or more processors, to cause the apparatus to (claim 25) having program code recorded thereon, the program code executed by one or more processors and comprising (mere instructions to apply an exception, (2) Whether the claim invokes computers - MPEP 2106.05(f); generic computer components)
observing an environment via one or more sensors associated with a robotic device (insignificant extra-solution activity – MPEP 2106.05(g), (3) observing an environment is data gathering and outputting; using sensors associated with a robotic device is mere instruction to apply the exception using generic computer components)
via an inference model associated with the robotic device (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; using a model to generate a belief)
performing, by the robotic device, an action in the environment based on the belief and the observation of the environment (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; perform an action)
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is directed to an abstract idea.
2B: The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
(claim 17) one or more processors; and one or more memories coupled with the one or more processors and storing instructions operable, when executed by the one or more processors, to cause the apparatus to (claim 25) having program code recorded thereon, the program code executed by one or more processors and comprising (mere instructions to apply an exception, (2) Whether the claim invokes computers - MPEP 2106.05(f); generic computer components)
observing an environment via one or more sensors associated with a robotic device (insignificant extra-solution activity – MPEP 2106.05(g), (3) data gathering and outputting, and WURC: receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 - MPEP 2106.05(d)(II)(i)); via one or more sensors associated with a robotic device is mere instruction to apply exception using generic computer components)
via an inference model associated with the robotic device (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; using a model to generate a belief)
performing, by the robotic device, an action in the environment based on the belief and the observation of the environment (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; perform an action)
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
With respect to claims 2, 10, 18 and 26:
2A Prong 2: The judicial exception is not integrated into a practical application.
further comprising training the inference model based on the data associated with prior actions of an expert in the environment (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; training the model based on the data)
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is directed to an abstract idea.
2B: The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
further comprising training the inference model based on the data associated with prior actions of an expert in the environment (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; training the model based on the data)
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
With respect to claims 3, 11, 19 and 27:
2A Prong 2: The judicial exception is not integrated into a practical application.
wherein a dynamics model trains the inference model (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; using a dynamics model training the inference model)
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is directed to an abstract idea.
2B: The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
wherein a dynamics model trains the inference model (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; using a dynamics model training the inference model)
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
With respect to claims 4, 12, 20 and 28:
2A Prong 2: The judicial exception is not integrated into a practical application.
wherein the data associated with the prior actions is deconfounded from the inference model (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; using an inference mode to deconfound the data associated with the prior actions; in light of spec [0051] ‘perform actions based on the expert data that is deconfounded with the inference model.’)
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is directed to an abstract idea.
2B: The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
wherein the data associated with the prior actions is deconfounded from the inference model (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; using an inference mode to deconfound the data associated with the prior actions)
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
With respect to claims 5, 13, 21 and 29:
2A Prong 2: The judicial exception is not integrated into a practical application.
wherein the inference model is a component of a variational encoder-decoder (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; claim 2 recites “training the inference model”; training a VAE model)
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is directed to an abstract idea.
2B: The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
wherein the inference model is a component of a variational encoder-decoder (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; claim 2 recites “training the inference model”; training a VAE model)
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
With respect to claims 6, 14, 22 and 30:
2A Prong 1: The claim recites a judicial exception.
wherein training the inference model includes minimizing a first loss for the dynamic model and minimizing a second loss for the inference model (mathematical concept - mathematical equation, in light of spec [0056])
With respect to claims 7, 15 and 23:
2A Prong 2: The judicial exception is not integrated into a practical application.
further comprising observing the prior actions of the agent in the environment via one or more sensors of the robotic device (insignificant extra-solution activity – MPEP 2106.05(g), (3) data gathering and outputting; using sensors to receive observation data including prior actions)
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is directed to an abstract idea.
2B: The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
further comprising observing the prior actions of the agent in the environment via one or more sensors of the robotic device (insignificant extra-solution activity – MPEP 2106.05(g), (3) data gathering and outputting, and WURC: receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 - MPEP 2106.05(d)(II)(i)); using sensors to receive observation data including prior actions)
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
With respect to claims 8, 16 and 24:
2A Prong 2: The judicial exception is not integrated into a practical application.
wherein the expert is a human or another robotic device (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; claim 2 recites “training the inference model based on the data associated with prior actions of an expert,” which is mere instructions to apply an exception, specifying more details about the data does not cause the limitation to integrate the exception into a practical application)
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is directed to an abstract idea.
2B: The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
wherein the expert is a human or another robotic device (mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception; claim 2 recites “training the inference model based on the data associated with prior actions of an expert,” which is mere instructions to apply an exception, specifying more details about the data does not cause the limitation to be significantly more than the judicial exception)
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
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 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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-30 rejected under 35 U.S.C. 103 as being unpatentable over Rafailov ("Visual Adversarial Imitation Learning using Variational Models" 20220627) in view of Yavuz ("Design of a String Encoder-and-IMU-Based 6D Pose Measurement System for a Teaching Tool and Its Application in Teleoperation of a Robot Manipulator" 20210616)
In regard to claims 1, 9, 17 and 25, Rafailov teaches: A processor-implemented method comprising: (Rafailov, p. 7, "All experiments were carried out on a single Titan RTX GPU using an internal cluster for about 1000 GPU hours.")
… generating, via an inference model associated with the robotic device, a belief of the environment based on the observation of the environment and data associated with prior actions of the robotic device in the environment, (Rafailov, p. 7, Environments and Demonstration Data "These consist of two locomotion environments from the DeepMind Control Suite [30], the classic Car Racing environment from OpenAI Gym [31] and two dexterous manipulation tasks using the D’Claw [32] and Shadow Hand platforms. [the robotic device]"; p. 4, 3.2 "we consider learning a latent representation of the history zt = q(ht) [a belief zt]... Consider a POMDP, and let zt be a latent space representation of the history and belief state [a belief zt] such that P(st|x≤t, a<t) = P(st|zt)."; p. 5, Algorithm 1 "7: Estimate latent state from the belief distribution zt ~ qθ(·|x_t, z_t-1, a_t-1)... 16: Infer expert latent states z_E_1:T ~ qθ(·|x_E_1:T, a_E_1:T-1) [a belief z based on the observation x and data associated with prior actions a] using the belief model qθ [an inference model qθ]"; p. 15, B Practical Algorithm with Variational Model "We can introduce the belief distribution q(z_1:T|x_1:T, a_1:T-1)... To estimate the expectation, we can use sequential sampling from the belief distribution zt ~ qθ(·|x_t, z_t-1, a_t-1), t = 1 : T and the reparameterization trick [52]…"; p. 6, 3.3 Practical Algorithm with Variational Models "where qθ is a state inference network, [an inference model qθ] Tθ is a latent dynamics model"; zt is an internal "belief" over the environment's state using past data)
the belief representing information of the environment not directly observable by the one or more sensors; and (Rafailov, p. 4, 3.2 Extension to POMDPs "In POMDPs, the underlying state is not directly observed [the belief representing information not directly observable], and thus cannot be directly used by the policy... By using the historical information, the belief state provides more information about the current state, and can enable the learning of better policies.")
performing, by the robotic device, an action in the environment based on the belief and the observation of the environment. (Rafailov, p. 2, 2 Preliminaries "We consider the problem setting of learning in partially observed Markov decision processes (POMDPs), which can be described with the tuple: M = (S, A, X, R, T, U, r), where… a ∈ A is the action space, x ∈ X is the observation space… The state evolution is Markovian and governed by the dynamics as s' ~ T(·|s, a)... The agent can interact with the environment and must learn a policy π(at|x≤t) [performing an action a based on the observation x] that mimics the expert."; p. 5, Algorithm 1 "7: Sample action at~ πψ(at|zt) [based on the belief z]")
Rafailov does not teach, but Yavuz teaches: observing an environment via one or more sensors associated with a robotic device; (Yavuz, p. 3, 2.2. Sensors and Infrastructure Used "The overall system [a robotic device] setup includes a robot, an IMU, six string encoder sensors, two designed system controller boards, one designed system program, and one personal computer (PC)."; p. 2, 1. Introduction "In this study, a sensor system that consists of an IMU and six string encoder position sensors [sensors] to measure and record the 6D pose of a teaching tool which is controlled by an operator during a teaching process is proposed... six string encoder position sensors are used to locate the 3D position of the tip of the teaching tool and to find the yaw angle of the tool (rotation around the z-axis) by using the law of cosines.")
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified Rafailov to incorporate the teachings of Yavuz by including IMU and string encoder sensors. Doing so would allow the use of those sensors to track exact physical movement. Specifically, it can locate the 3D position of the tip of the teaching tool and find the yaw angle of the tool (rotation around the z-axis) by using the law of cosines. (Yavuz, p. 2, 1. Introduction "... six string encoder position sensors are used to locate the 3D position of the tip of the teaching tool and to find the yaw angle of the tool (rotation around the z-axis) by using the law of cosines.")
Claims 9, 17 and 25 recite substantially the same limitation as claim 1, therefore the rejection applied to claim 1 also apply to claims 9, 17 and 25. In addition, Rafailov teaches: (claim 17) one or more processors; and one or more memories coupled with the one or more processors and storing instructions operable, when executed by the one or more processors, to cause the robotics device to (claim 25) having program code recorded thereon, the program code executed by one or more processors and comprising (Rafailov, p. 7, "All experiments were carried out on a single Titan RTX GPU using an internal cluster for about 1000 GPU hours.")
In regard to claims 2, 10, 18 and 26, Rafailov teaches: further comprising training the inference model based on the data associated with prior actions of an expert in the environment. (Rafailov, p. 5, Algorithm 1 "1: Require : Expert demos BE, environment buffer Bπ… 12: Sample a batch of trajectories {x_1:T, a_1:T-1} from the joint buffer BE [the data associated with prior actions of an expert] ∪ Bπ 13: Optimize the variational model {qθ, Τ^θ} using Equatin 7 [training the inference model qθ]"; p. 6, 3.3 Practical Algorithm with Variational Models "where qθ is a state inference network, Tθ is a latent dynamics model")
In regard to claims 3, 11, 19 and 27, Rafailov teaches: wherein a dynamics model trains the inference model. (Rafailov, p. 5, 3.3 Practical Algorithm with Variational Models "The divergence bound of Theorem 1 allows us to develop a practical algorithm if we can learn a good belief state representation. Following prior work [23, 24, 25, 26, 27, 28] we optimize the ELBO: max E_qθ[... logU(xt|zt) - D_KL(qθ(z_t|x_t, z_t-1, a_t-1))||Τθ(z_t|z_t-1, a_t-1))] (7) where qθ is a state inference network, Tθ is a latent dynamics model...Here we jointly train a belief representation with network qθ and a latent dynamics model T^θ. [the dynamic model Tθ trains the inference model qθ]"; the two models learn at the same time by minimizing the KL divergence between them, maximizing the negative KL divergence is minimizing the KL divergence, which helps the two models align)
In regard to claims 4, 12, 20 and 28, Rafailov teaches: wherein the data associated with the prior actions is deconfounded from the inference model. (Rafailov, p. 5, Algorithm 1 "15: Sample trajectories from expert buffer {x_E_1:T, a_E_1:T-1[the data associated with the prior actions]} ~BE 16: Infer expert latent states z_E_1:T [deconfounded data, hidden data] ~ qθ(·|x_E_1:T, a_E_1:T-1) using the belief model qθ"; in light of spec [0051] 'perform actions based on the expert data that is deconfounded with the inference model... the effect of hidden information on the expert and then account for the effect of the hidden information once the inference model has been trained)
In regard to claims 5, 13, 21 and 29, Rafailov teaches: wherein the inference model is a component of a variational encoder-decoder. (Rafailov, p. 16, B Practical Algorithm with Variational Models "We base our network architectural choice on the recurrent state space model [a variational encoder-decoder] [27, 28], as it has shown strong performance in RL tasks from images."; p. 5, 3.2 Extension to POMDPs "If we can learn an encoder zt = q(x≤t, a<t) [a component of a variational encoder-decoder.] that captures sufficient statistics of the history"; the state inference network qθ acts as an encoder within a recurrent state-space variational model. It functions alongside a latent dynamics model Tθ and an observation model U, which together form a variational encoder-decoder architecture)
In regard to claims 6, 14, 22 and 30, Rafailov teaches: wherein training the inference model includes minimizing a first loss for the dynamic model and minimizing a second loss for the inference model. (Rafailov, p. 5, 3.3 Practical Algorithm with Variational Models "The divergence bound of Theorem 1 allows us to develop a practical algorithm if we can learn a good belief state representation. Following prior work [23, 24, 25, 26, 27, 28] we optimize the ELBO: max E_qθ[... logU(xt|zt) - D_KL(qθ(z_t|x_t, z_t-1, a_t-1))||Τθ(z_t|z_t-1, a_t-1))] (7) where qθ is a state inference network, Tθ is a latent dynamics model [minimizing a first loss for the dynamic model Tθ and minimizing a second loss for the inference model qθ]"; p. 16, B Practical Algorithm with Variational Models "This leads to the empirical model loss: max E_qθ[... logU(xt|zt) - D_KL(qθ(z_t|x_t, z_t-1, a_t-1))||Τθ(z_t|z_t-1, a_t-1))]. (21) That is, we jointly train a belief representation qθ and the Markovian dynamics model Tθ"; maximized a loss function is minimized its negative, i.e. loss function * -1)
In regard to claims 7, 15 and 23, Rafailov does not teach, but Yavuz teaches: further comprising observing the prior actions of the agent in the environment via one or more sensors of the device. (Yavuz, p. 3, 2.2. Sensors and Infrastructure Used "The overall system [the robotic device] setup includes a robot, an IMU, six string encoder sensors, two designed system controller boards, one designed system program, and one personal computer (PC)."; p. 2, 1. Introduction "In this study, a sensor system that consists of an IMU and six string encoder position sensors [sensors of the robotic device] to measure and record the 6D pose of a teaching tool which is controlled by an operator [prior actions of the agent/a human] during a teaching process is proposed... six string encoder position sensors are used to locate the 3D position of the tip of the teaching tool and to find the yaw angle of the tool (rotation around the z-axis) by using the law of cosines. ")
The rationale for combining the teachings of Rafailov and Yavuz is the same as set forth in the rejection of claim 1.
In regard to claims 8, 16 and 24, Rafailov teaches: wherein the expert is a human or another robotic device. (Rafailov, p. 5, Algorithm 1 "1: Require : Expert demos BE, [demonstrations from a human or an expert] environment buffer Bπ."; p. 6, 3.4 Zero-Shot Transfer to New Imitation Tasks "we first train on all of the source tasks using Algorithm 1, training a single shared variational model across the tasks. By fine-tuning that model on data that includes the target task expert demonstrations our hope is that we can get an accurate model and thus a high-quality policy.")
Response to Arguments
Applicant's amendments with respect to the claim objections have been fully considered and are sufficient to overcome the objections. The objections have been withdrawn.
Applicant's amendments regarding the § 112(b) rejection have been fully considered, but the rejection is maintained in view of the newly added limitations.
Applicant's arguments with respect to the rejection of the claims under 35 U.S.C. 101 have been fully considered but they are not persuasive:
Argument: (p. 8-9) However, the claims recite operations that cannot be performed in the human mind. For example, claim 1 recites, in part, "generating, via an inference model associated with the robotic device, a belief of the environment." Here, the inference model is a computational construct (as described in the specification at [0054], [0061]) that processes sensor data through neural network architectures to generate probability distributions over estimated hidden information. This is not a mental process performable by a human. Claim 1 also recites, "the belief representing information of the environment not directly observable by the one or more sensors." The belief represents hidden variables (0) that are inferred computationally from sequences of observations and actions.
Response: the claim recites “generating… a belief of the environment based on the observation of the environment and data associated with prior actions…,” which describes pure data manipulation, or estimating hidden states from observations and prior actions. This type of estimation or data processing can be performed in the human mind. Further, the recitation of “via an inference model associated with the robotic device” is mere instruction to apply the exception using generic computer components.]
Argument: (p. 9) Claim 1 further recites, "performing, by the robotic device, an action in the environment based on the belief and the observation of the environment." This limitation alone removes the claims from the realm of mental processes. A human cannot perform the actions of a robotic device.
Response: the performing step is not an abstract idea, it is identified as an additional element as mere instructions to apply an exception – MPEP 2106.05(f), (3) The particularity or generality of the application of the judicial exception.
Argument: (p. 10-11) Rather, claim 1 is directed to a technical solution to a technical problem, namely enabling a robotic device to perform tasks in environments where critical information affecting optimal action selection is not directly observable by the robot's sensors… This is a concrete technological improvement to robotic control systems, not an abstract idea… Claim 1 improves the functioning of the robotic system itself by enabling it to act on inferred hidden information. As disclosed in the specification, the inference model learns to represent hidden variables that affect expert behavior but are not directly observable ([0042], [0044]), the system addresses the problem of causal confounding in imitation learning ([0042]), and the encoder generates probability distributions for estimated hidden information given observations and actions ([0054]).
Response: the claim recites “generating… a belief of the environment based on the observation of the environment and data associated with prior actions…,” which is identified as an abstract idea. If the claim is directed to a judicial exception, it cannot provide an improvement. See MPEP 2106.05(a): “It is important to note, the judicial exception alone cannot provide the improvement.” Further, “the belief representing information of the environment not directly observable by the one or more sensors” as part of the mental process that can be performed mentally as a result of evaluation, opinion, and judgment.]
Argument: (p. 11) Furthermore, claim 1 transforms sensor observations and action histories into inferred beliefs, which then control physical robotic actions. This is not insignificant post-solution activity-the physical manipulation of the environment by the robotic device is the very purpose of the invention.
Response: the generating step is an abstract idea as a mental process, and the performing step is an additional element as mere instructions to apply an exception (MPEP 2106.05(f)). Both steps are not additional elements as insignificant extra-solution activity (MPEP 2106.05(g)).
Argument: (p. 11-12) Here, claim 1 improves robotic control technology by enabling robotic devices to perform tasks that require acting on information not directly observable by sensors. As disclosed in the specification at [0027], [0042], conventional robotic imitation learning fails when the expert demonstrator has access to hidden information that the robot cannot sense. The claimed aspects solve this technical problem by computationally inferring the hidden information via an inference model that processes both observations and prior actions to generate a belief representing the non-observable information. This is not invoking a computer merely as a tool to perform an existing process. Rather, claim 1 recites a specific improvement to robotic control systems that enables robots to successfully perform tasks in environments with hidden variables-a capability that conventional sensor-based robotic systems lack.
Response: The applicant states the improvement uses hidden variables to enable successful robot performance; however, what the claim recites regarding the hidden variables is ‘generating… a belief of the environment,” which is an abstract and therefore is not sufficient to provide an improvement. If the claim is directed to a judicial exception, it cannot provide an improvement. See MPEP 2106.05(a): “It is important to note, the judicial exception alone cannot provide the improvement.”
Argument: (p. 12) The combination of elements is unconventional. Claim 1 recites using an inference model to generate beliefs representing hidden environmental information, basing the belief on both current observations and prior action data, and controlling a robotic device based on both the inferred belief and direct observations… The Examiner has provided no evidentiary support for the assertion that this ordered combination of elements is well-understood, routine, and conventional in the field of robotic control. The ordered combination of these elements amounts to significantly more than any alleged abstract idea.
Response: Applicant appears to argue that the ordered combination amounts to significantly more. The claim recites “generating… a belief… based on the observation… and data associated with prior actions” to “an action… based on the belief and the observation of the environment,” which is a combination of abstract ideas. When incorporating the exceptions with the additional elements to evaluate the claim as a whole, the ordered combination of elements of mental steps remains a series of mental steps, i.e. the combination of mental steps does not make it patent-eligible.
Applicant's arguments with respect to the rejection of the claims under 35 U.S.C. 103 have been fully considered but they are moot:
Argument: (p. 13-15) As amended, claim 1 recites, in part, "generating, via an inference model associated with the robotic device, a belief of the environment based on the observation of the environment and data associated with prior actions of the robotic device in the environment, the belief representing information of the environment not directly observable by the one or more sensors."… Pertsch describes an internal representation of the agent's selected action strategy, not a belief about an environmental property. Claim 1, by contrast, recites a belief representing information of the environment itself, wherein "the belief [represents] information of the environment not directly observable by the one or more sensors."
Response: the arguments do not apply to the reference (Rafailov) being used in the current rejection.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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 SU-TING CHUANG whose telephone number is (408)918-7519. The examiner can normally be reached Monday - Thursday 8-5 PT.
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, Usmaan Saeed can be reached at (571) 272-4046. 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.
/S.C./Examiner, Art Unit 2146
/SHAHID K KHAN/Primary Examiner, Art Unit 2146
Prosecution Insights