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 Status
The status of claims 1-16 is:
Claims 1-16 were pending as of the Non-Final Rejection mailed 09/25/2025.
Claims 1, 7-11, and 13-14 are amended as of the amendments and remarks received 03/24/2026.
Claims 2-6, 12, and 15-16 remain as originally presented as of the amendments and remarks received 03/24/2026.
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.
35 U.S.C. 101 requires that a claimed invention must fall within one of the four eligible categories of invention (i.e. process, machine, manufacture, or composition of matter) and must not be directed to subject matter encompassing a judicially recognized exception as interpreted by the courts. MPEP 2106. Three categories of subject matter are found to be judicially recognized exceptions to 35 U.S.C. § 101 (i.e. patent ineligible) (1) laws of nature, (2) physical phenomena, and (3) abstract ideas. MPEP 2106(II). To be patent-eligible, a claim directed to a judicial exception must as whole be integrated into a practical application or directed to significantly more than the exception itself (MPEP 2106). Hence, the claim must describe a process or product that applies the exception in a meaningful way, such that it is more than a drafting effort designed to monopolize the exception.
Claims 1-6, 9-10, and 14-16 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Independent claims 1 and 14 are directed to one of the four statutory categories of eligible subject matter; thus, the claims pass Step 1 of the Subject Matter Eligibility Test (See flowchart in MPEP 2106).
Step 2A, Prong 1 Analysis
Independent claim 1 is directed to in a robot equipped with one or more camera-based object detectors, receiving an input, the input comprising point cloud observations, from the one or more camera-based object detectors, of a local region and localization of a robot camera pose of the robot; and outputting a viewpoint for the robot to move to as a result of sequential online planning; and directing the robot to move to the outputted view. An individual can receive an input comprising observations of a region and determine a viewpoint to move to and direct a robot to move to that viewpoint. The collection of point cloud observations and localization of a robot camera pose are insignificant data acquisition. Accordingly, the analysis under prong one of Step 2A of the Subject Matter Eligibility Test does not result in a conclusion of eligibility (See flowchart in MPEP 2106).
Independent claim 14 is directed to a system comprising: a robot equipped with one or more camera-based object detectors; and a gRPC framework comprising a gRPC client and a gRPC server, the gRPC client providing an interface between the robot and the gRPC server, the gRPC server maintaining an occupancy octree, a Partially Observable Markov Decision Process (POMDP) agent and a belief state; wherein the gRPC client directs the robot to move to a viewpoint determined by the POMDP agent. Maintaining an occupancy octree, a Partially Observable Markov Decision Process (POMDP) agent and a belief state are all mathematical concepts. Further, an individual could direct a robot to move to a determined viewpoint. Accordingly, the analysis under prong one of Step 2A of the Subject Matter Eligibility Test does not result in a conclusion of eligibility (See flowchart in MPEP 2106).
Additional elements
Independent claim 1 claims a robot equipped with one or more camera-based object detectors. Independent claim 14 claims a robot equipped with one or more camera-based object detectors; and a gRPC framework comprising a gRPC client and a gRPC server.
Step 2A, Prong 2 Analysis
The above-identified elements do not integrate the judicial into a practical application nor do they suggest an improvement.
The additional elements of a robot or automated machine equipped with one or more camera-based object detectors and a gRPC framework comprising a gRPC client and a gRPC server amounts to merely using generic computer hardware or components as a tool to perform the claimed mental process.
Using a general purpose computer to apply a judicial exception does not qualify as a particular machine and therefore, does not integrate a judicial exception into a practical application (See MPEP 2106.05(b)). Furthermore, implementing an abstract idea on a computer does not integrate a judicial exception into a practical application (See MPEP 2106.05(f)).
Moreover, the additional elements of the claims do not recite an improvement in the functioning of a computer or another technology or technical field, the claimed steps do not effect a transformation, and the claims do not apply the judicial exception in any meaningful way beyond generically linking the use of the judicial exception to a particular technological environment (See MPEP 2106.04(d)).
Further, the act of acquiring data is mere data gathering which amounts to insignificant extra-solution activity (See MPEP 2106.05(g)). Therefore, the analysis under prong two of step 2A of the Subject Matter Eligibility Test does not result in a conclusion of eligibility (See flowchart in MPEP 2106).
Step 2B
Finally, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Regarding independent claims 1 and 16, as noted above, the additional elements are generic computer features which perform generic computer functions that are well-understood, routine, and conventional and do not amount to more than implementing the abstract idea with a computerized system. Thus, taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception (the abstract idea).
Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves and other technology. Their collective functions merely provide conventional computer implementation, and mere implementation on a generic computer does not add significantly more to the claims. Accordingly, the analysis under step 2B of the Subject Matter Eligibility Test does not result in a conclusion of eligibility (See flowchart in MPEP 2106).
For all the foregoing reasons, independent claims [] do not recite eligible subject matter under 35 USC 101.
Claim 2 claims wherein the input further comprises three dimensional (3D) bounding boxes with detected object labels. The features of claim 2 are directed to the mental process since they do not preclude the mental analysis as recited in claim 1. Accordingly, claim 2 does not integrate the judicial exception into a practical application or amount to significantly more than the judicial exception.
Claim 3 claims wherein each of the object labels comprises a label. The features of claim 3 are directed to the mental process since they do not preclude the mental analysis as recited in claim 1. Accordingly, claim 3 does not integrate the judicial exception into a practical application or amount to significantly more than the judicial exception.
Claim 4 claims wherein the input further comprises segmented point clouds for detected objects with detected object labels. The features of claim 4 are directed to the mental process since they do not preclude the mental analysis as recited in claim 1. Accordingly, claim 4 does not integrate the judicial exception into a practical application or amount to significantly more than the judicial exception.
Claim 5 claims wherein the input further comprises two dimensional (2D) bounding boxes on an image paired with a corresponding depth image with detected object labels. The features of claim 5 are directed to the mental process since they do not preclude the mental analysis as recited in claim 1. Accordingly, claim 5 does not integrate the judicial exception into a practical application or amount to significantly more than the judicial exception.
Claim 6 claims wherein the input further comprises detected object labels. The features of claim 6 are directed to the mental process since they do not preclude the mental analysis as recited in claim 1. Accordingly, claim 6 does not integrate the judicial exception into a practical application or amount to significantly more than the judicial exception.
Claim 9 claims wherein determining viewpoints for the robot to move to and observe at is performed by sequential decision-making based on Partially Observable Markov Decision Process (POMDP) model for three dimensional multi-object search. The features of claim 9 are further mathematical processes. Accordingly, claim 9 does not integrate the judicial exception into a practical application or amount to significantly more than the judicial exception.
Claim 10 claims wherein viewpoint candidates are initialized and updated by sampling from the local region based on a current information state and occupancy to form a viewpoint graph. The features of claim 10 are further mathematical processes. Accordingly, claim 10 does not integrate the judicial exception into a practical application or amount to significantly more than the judicial exception.
Claim 15 claims wherein the belief state represents belief over object locations in the structure of the occupancy octree. The features of claim 15 are directed to the mental process since they do not preclude the mental analysis as recited in claim 1. Accordingly, claim 15 does not integrate the judicial exception into a practical application or amount to significantly more than the judicial exception.
Claim 16 claims wherein the occupancy octree represents a search region's occupancy. The features of claim 16 are directed to the mental process since they do not preclude the mental analysis as recited in claim 1. Accordingly, claim 16 does not integrate the judicial exception into a practical application or amount to significantly more than the judicial exception.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1, 4, 6-8, and 11 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Patten et al. (Patten, T., Zillich, M., Fitch, R., Vincze, M., & Sukkarieh, S. (2015). Viewpoint evaluation for online 3-D active object classification. IEEE Robotics and Automation Letters, 1(1), 73-81., hereinafter “Patten”).
Regarding claim 1, Patten discloses a method comprising:
in a robot equipped with one or more camera-based object detectors (Patten Page 29: “Our robot is a Festo Robotino with custom-mounted ASUS XTion Pro Live RGB-D sensor”), receiving an input (Patten Page 74: “The robot processes point cloud observations captured with an onboard sensor and selects the next location to make an observation”), the input comprising point cloud observations, from the one or more camera-based object detectors, of a local region (Patten Page 74: “The robot processes point cloud observations captured with an onboard sensor and selects the next location to make an observation”), and localization of a robot camera pose of the robot (Patten Page 74: “From location yk, the robot makes a point cloud observation Zk and stores it in a global 3D occupancy grid G”, implies robot camera pose localization because the camera would not be able to move to another viewpoint if it did not know its current pose and location); and
outputting a viewpoint for the robot to move to as a result of sequential online planning (Patten Pages 74-75: “The set of candidate viewpoints is Yk = Y0\y1:k, where Y0 is the initial set of available locations and y1:k = {y1,y2,...,yk} is the history of visited locations. The next location is chosen as the one which yields the largest utility. The active classification problem is defined as follows. Given observations Z1:k = {Z1,Z2,...,Zk}made from visited locations y1:k and the current belief Bk at stage k, choose the next location yk+1 from the set of available locations Yk by maximising the utility function”); and
directing the robot to move to the outputted viewpoint (Patten Pages 74-75: “The set of candidate viewpoints is Yk = Y0\y1:k, where Y0 is the initial set of available locations and y1:k = {y1,y2,...,yk} is the history of visited locations. The next location is chosen as the one which yields the largest utility. The active classification problem is defined as follows. Given observations Z1:k = {Z1,Z2,...,Zk}made from visited locations y1:k and the current belief Bk at stage k, choose the next location yk+1 from the set of available locations Yk by maximising the utility function”).
Regarding claim 4, Patten discloses the method wherein the input further comprises segmented point clouds for detected objects with detected object labels (Patten Page 76: “Each observation is stored in a 3D occupancy grid G. The occupied space of each object n can be computed by determining the voxels that correspond to the observations zn1:k−1. The set of voxels for each new segment is compared to the voxel set of each object in the prior belief Nk−1. A segment is associated to an object if the proportion of voxels overlapping those of a prior object is greater than a given threshold. The proportion is computed as the sum of the number of overlapping voxels divided by the total number of occupied voxels for the segment. Set Nk is maintained by adding new objects for unassociated segments and merging associated segments with existing objects as appropriate”; Patten Page 77: “The descriptors for each viewpoint and instance, along with the class label, are stored in a k-d tree with dimension 640").
Regarding claim 6, Patten discloses the method wherein the input further comprises detected object labels (Patten Page 77: “Classification is performed using [22]. The classifier is built offline by generating partial point clouds of object models in a pre-defined database. The database consists of a large number of model instances which are grouped into classes. Partial point clouds are generated from given locations on a 3D view sphere and a global feature descriptor is computed for each point cloud. Here, we use the ensemble of shape functions (ESF) descriptor, which consists of ten 64-bin histograms based on distinct shape functions (distance, angle and area distributions) [23], although any global feature descriptor could be used. The descriptors for each viewpoint and instance, along with the class label, are stored in a k-d tree with dimension 640").
Regarding claim 7, Patten discloses the method further comprising maintaining information states regarding target object locations through a probability distribution structured as an octree (Patten Page 79: “Software is written in C++ using the point cloud library [24], octomap [25] and ROS [26]”) and updated based on object detection observations or the point cloud observations (Patten Page 73: “We maintain a set of object hypotheses representing class and pose estimates that are incrementally updated with each new observation”).
Regarding claim 8, Patten discloses the method further comprising:
dynamically determining occupancy of a search region through constructing an octree-based occupancy grid based on point cloud observations (Patten Page 74: “From location yk, the robot makes a point cloud observation Zk and stores it in a global 3D occupancy grid G”; Patten Page 79: “Software is written in C++ using the point cloud library [24], octomap [25] and ROS [26]”); and
using ray-tracing to determine visibility at three dimensional locations within the local region (Patten Page 73: “Online, we use ray tracing to predict which points of an object would be visible from a particular viewpoint given the other objects observed so far and thus compute a utility measure that indicates the discriminatory power of that viewpoint”).
Regarding claim 11, Patten discloses a method comprising:
in an automated machine equipped with one or more camera-based object detectors (Patten Page 29: “Our robot is a Festo Robotino with custom-mounted ASUS XTion Pro Live RGB-D sensor”), receiving human-provided information or information inferred from point cloud observations regarding target locations (Patten Page 74: “The robot processes point cloud observations captured with an onboard sensor and selects the next location to make an observation”);
maintaining information states regarding the target locations through a probability distribution structured as an octree (Patten Page 79: “Software is written in C++ using the point cloud library [24], octomap [25] and ROS [26]”);
initializing the information states based on point cloud observations (Patten Page 74: “From location yk, the robot makes a point cloud observation Zk and stores it in a global 3D occupancy grid G”);
updating the information states based on object detection observations or point cloud observations (Patten Page 73: “We maintain a set of object hypotheses representing class and pose estimates that are incrementally updated with each new observation”);
determining occupancy of a search region through constructing an octree-based occupancy grid based on the point cloud observations (Patten Page 74: “From location yk, the robot makes a point cloud observation Zk and stores it in a global 3D occupancy grid G”; Patten Page 79: “Software is written in C++ using the point cloud library [24], octomap [25] and ROS [26]”); and
using ray-tracing to determine visibility at three dimensional locations within the search region (Patten Page 73: “Online, we use ray tracing to predict which points of an object would be visible from a particular viewpoint given the other objects observed so far and thus compute a utility measure that indicates the discriminatory power of that viewpoint”); and
directing the automated machine to move to a viewpoint determined based on the information states (Patten Pages 74-75: “The set of candidate viewpoints is Yk = Y0\y1:k, where Y0 is the initial set of available locations and y1:k = {y1,y2,...,yk} is the history of visited locations. The next location is chosen as the one which yields the largest utility. The active classification problem is defined as follows. Given observations Z1:k = {Z1,Z2,...,Zk}made from visited locations y1:k and the current belief Bk at stage k, choose the next location yk+1 from the set of available locations Yk by maximising the utility function”).
Claim Rejections - 35 USC § 103
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) 2-3 and 5 are rejected under 35 U.S.C. 103 as being unpatentable over Patten in view of Robinson et al. (Robinson, B., Langford, D., Jetton, J., Cannan, L., Patterson, K., Diltz, R., & English, W. (2021, April). Real-time object detection and geolocation using 3D calibrated camera/LiDAR pair. In Autonomous Systems: Sensors, Processing, and Security for Vehicles and Infrastructure 2021 (Vol. 11748, pp. 57-77). SPIE., hereinafter “Robinson”).
Regarding claim 2, Patten discloses the method wherein the input further comprises three dimensional (3D) bounding boxes (Patten Page 76: “We speed up the process by first comparing the bounding boxes of the object point clouds in the two different sets. Voxel matching is then only performed with the objects that pass the quick bounding box check”, voxels implies the bounding boxes are 3D).
Patten does not explicitly disclose the method wherein the bounding boxes have detected object labels (although Patten does disclose object labels on Page 76: “Each observation is stored in a 3D occupancy grid G. The occupied space of each object n can be computed by determining the voxels that correspond to the observations zn1:k−1. The set of voxels for each new segment is compared to the voxel set of each object in the prior belief Nk−1. A segment is associated to an object if the proportion of voxels overlapping those of a prior object is greater than a given threshold. The proportion is computed as the sum of the number of overlapping voxels divided by the total number of occupied voxels for the segment. Set Nk is maintained by adding new objects for unassociated segments and merging associated segments with existing objects as appropriate”; Patten Page 77: “The descriptors for each viewpoint and instance, along with the class label, are stored in a k-d tree with dimension 640").
However, Robinson teaches bounding boxes with labels (Robinson Fig. 19: shows the bounding boxes with labels).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate having the class labels with the bounding box as taught by Robinson with the method of Patten because it would allow for bounding boxes to be displayed to a user that could see the physical space of an object detected and the class of object the method detected. This motivation for the combination of Patten and Robinson is supported by KSR exemplary rationale (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results. MPEP 2141 (III).
Regarding claim 3, Patten does not explicitly disclose the method wherein each of the object labels comprises a label.
However, Robinson teaches wherein each of the object labels comprises a label (Robinson Fig. 19: shows the bounding boxes with labels).
It would have been obvious to combine Patten and Robinson for the same reasons as used for claim 2 above.
Regarding claim 5, Patten does not explicitly disclose the method wherein the input further comprises two dimensional (2D) bounding boxes on an image paired with a corresponding depth image with detected object labels.
However, Robinson teaches the input further comprises two dimensional (2D) bounding boxes on an image (Robinson Fig. 19: shows the bounding boxes with labels) paired with a corresponding depth image (Robinson Fig. 18: shows a depth lidar image that is used to detect (paired) with objects that are detected in the other image) with detected object labels (Robinson Fig. 19: shows the bounding boxes with labels).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate an image with bounding boxes and class labels as taught by Robinson with the method of Patten because it would allow for bounding boxes to be displayed to a user that could see the area of an object detected and the class of object the method detected. This motivation for the combination of Patten and Robinson is supported by KSR exemplary rationale (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results. MPEP 2141 (III).
Claim(s) 9-10 and 12-13 are rejected under 35 U.S.C. 103 as being unpatentable over Patten in view of Eidenberger et al. (Eidenberger, R., & Scharinger, J. (2010, October). Active perception and scene modeling by planning with probabilistic 6d object poses. In 2010 IEEE/RSJ international conference on intelligent robots and systems (pp. 1036-1043). IEEE., hereinafter “Eidenberger”).
Regarding claim 9, Patten does not disclose the method further comprising determining additional viewpoints for the robot to move to and observe at by sequential decision-making based on a Partially Observable Markov Decision Process (POMDP) model for three dimensional multi-object search.
However, Eidenberger teaches the method further comprising determining additional viewpoints for the robot to move to and observe at by sequential decision-making based on a Partially Observable Markov Decision Process (POMDP) model for three dimensional multi-object search (Eidenberger Page 1: “To improve detection results and to tackle occlusion problems a method for active planning is proposed which reasons about model and state transition uncertainties in continuous and high dimensional domains. Information theoretic quality criteria are used for sequential decision making to evaluate probability distributions. The probabilistic planner is realized as a partially observable Markov decision process (POMDP)”).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the POMDP model as taught by Eidenberger with the method of Patten because it would improve the detection results as well as improve the method’s ability to handle occlusion problems (Eidenberger Page 1). This motivation for the combination of Patten and Eidenberger is supported by KSR exemplary rationale (G) Some teaching, suggestion, or motivation in the prior art that would have led one of ordinary skill to modify the prior art reference or to combine prior art reference teachings to arrive at the claimed invention and exemplary rational (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results. MPEP 2141 (III).
Regarding claim 10, Patten does not explicitly disclose the method, wherein further comprising initializing and updating viewpoint candidates by sampling from the local region based on a current information state and occupancy to form a viewpoint graph.
However, Eidenberger teaches the method, wherein further comprising initializing and updating viewpoint candidates by sampling from the local region based on a current information state and occupancy to form a viewpoint graph (Eidenberger Pages 4-5: “Therefore, samples are draw from the distributions and each sample arrangement is checked for intersections according to their object geometries”).
It would have been obvious to combine Patten and Eidenberger for the same reasons as used for claim 9 above.
Regarding claim 12, Patten does not explicitly disclose the method further comprising performing sequential decision-making based on a Partially Observable Markov Decision Process (POMDP) for three dimensional multi-object search to determine various viewpoints for the automated machine to move to and observe at.
However, Eidenberger teaches the method further comprising performing sequential decision-making based on a Partially Observable Markov Decision Process (POMDP) for three dimensional multi-object search to determine various viewpoints for the automated machine to move to and observe at (Eidenberger Page 1: “To improve detection results and to tackle occlusion problems a method for active planning is proposed which reasons about model and state transition uncertainties in continuous and high dimensional domains. Information theoretic quality criteria are used for sequential decision making to evaluate probability distributions. The probabilistic planner is realized as a partially observable Markov decision process (POMDP)”).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the POMDP model as taught by Eidenberger with the method of Patten because it would improve the detection results as well as improve the method’s ability to handle occlusion problems (Eidenberger Page 1). This motivation for the combination of Patten and Eidenberger is supported by KSR exemplary rationale (G) Some teaching, suggestion, or motivation in the prior art that would have led one of ordinary skill to modify the prior art reference or to combine prior art reference teachings to arrive at the claimed invention and exemplary rational (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results. MPEP 2141 (III).
Regarding claim 13, Patten does not explicitly disclose the method further comprising, signaling when an object is found, wherein a location of the found object is indicated in the information states at the time of the signaling.
However, Eidenberger teaches the method further comprising, signaling when an object is found, wherein a location of the found object is indicated in the information states at the time of the signaling (Eidenberger Page 6: “Initially we do not have any scene knowledge, so each action promises identical benefits. We start from the current robot position, namely viewpoint 6. After performing the first measurement and accomplishing the data association and state estimation process we retrieve the probabilistic scene distribution consisting of a total of 7 detected object instances, 3 of them of unsatisfactory pose accuracy, though. Table I lists the recognition results and the number of mixture components of the probability distributions during the state update. . . The first observation is performed at viewpoint 6. The salt box and the tomato soup can are recognized very well. The recognition results of the stapled soup boxes, especially for the lower blue one, are worth looking at. Two mixture components, one for the green and one for the blue box, are assigned to the object instance of the bottom soup box. This effect results from the similarity of the objects as they are almost identical in their textures, implying they have many similar interest points. For the Amicelli box in the front- due to reflections- and for some back objects no hypotheses are found as they are too far away or beyond the image. The planning algorithm aims at differentiating between the soup boxes and sharpening the knowledge of the Ceylon tee and the jam tins. It determines moving to viewpoint 5 as best future action. The planning results are shown in Table II. Due to the heavy occlusion of the bottom soup box, many viewpoints are considered as disadvantageous”).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate signaling when an object is detected as taught by Robinson with the method of Patten because it would improve the method by notifying a user of when and where an object has been detected. This motivation for the combination of Patten and Eidenberger is supported by KSR exemplary rationale (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results. MPEP 2141 (III).
Claim(s) 14-16 are rejected under 35 U.S.C. 103 as being unpatentable over Robinson in view of Vasquez-Gomez et al. (Vasquez-Gomez, J. I., Sucar, L. E., & Murrieta-Cid, R. (2017). View/state planning for three-dimensional object reconstruction under uncertainty. Autonomous Robots, 41(1), 89-109., hereinafter “Vasquez”), Ahmad et al. (Ahmad, S., Sunberg, Z. N., & Humbert, J. S. (2021). End-to-end probabilistic depth perception and 3d obstacle avoidance using pomdp. Journal of Intelligent & Robotic Systems, 103(2), 33., hereinafter “Ahmad”), and Patten.
Regarding claim 14, Robinson discloses a system comprising:
a robot equipped with one or more camera-based object detectors (Robinson Page 5: “The current system employs a camera, a LiDAR, an inertial measurement unit (IMU), and a global positioning system (GNSS) to provide near real-time detection and geolocation of artifacts on the runway surface”); and
a gRPC framework comprising a gRPC client and a gRPC server (Robinson Page 8: “Communication between the POS and the controller software would be performed using gRPC remote procedure calls (gRPC), with protocol buffers as the interface description language”), the gRPC client providing an interface between the robot and the gRPC server (Robinson Page 8: “Communication between the POS and the controller software would be performed using gRPC remote procedure calls (gRPC), with protocol buffers as the interface description language”).
Robinson does not explicitly disclose the system comprising:
the gRPC server maintaining an occupancy octree.
However, Vasquez teaches disclose the system comprising:
the gRPC server maintaining an occupancy octree (Vasquez Page 5: “After each scan, the sensor readings are integrated into an octree that represents the object’s bounding box”).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the octree structure as taught by Vasquez with the system of Robinson because it would improve the system by reducing positioning error, collision rate, and increasing the coverage (Vasquez Page 1). This motivation for the combination of Robinson and Vasquez is supported by KSR exemplary rationale (G) Some teaching, suggestion, or motivation in the prior art that would have led one of ordinary skill to modify the prior art reference or to combine prior art reference teachings to arrive at the claimed invention and exemplary rational (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results. MPEP 2141 (III).
The Robinson and Vasquez combination does not explicitly disclose the system comprising:
the gRPC server maintaining a Partially Observable Markov Decision Process (POMDP) agent and a belief state.
However, Ahmad teaches the system comprising:
the gRPC server maintaining a Partially Observable Markov Decision Process (POMDP) agent and a belief state (Ahmad Page 8: “At each time step, N particles are drawn at random according to the belief distribution bt . Each particle is propagated forward in time by drawing a sample from the importance distribution P (st+1 | st). An image received at the time step t + 1 is discretized and each voxel is assigned the number of image points contained within it. Each propagated particle is then assigned a weight according to the observation likelihood P (ot+1 | st+1) using Eqs. 2 and 5”).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate POMDP as taught by Ahmad with the system of Robinson and Vasquez because it would improve the system by allowing for object reconstruction that takes into account the uncertainty of reaching the state and the uncertainty in the observations of the sensors (Vasquez Page 21). This motivation for the combination of Robinson, Vasquez, and Ahmad is supported by KSR exemplary rationale (G) Some teaching, suggestion, or motivation in the prior art that would have led one of ordinary skill to modify the prior art reference or to combine prior art reference teachings to arrive at the claimed invention and exemplary rational (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results. MPEP 2141 (III).
The Robinson, Vasquez, and Ahmad combination does not explicitly disclose the system comprising:
wherein the gRPC client directs the robot to move to a viewpoint determined by the POMDP agent.
However, Patten teaches directing a robot to move to a determined viewpoint (Patten Pages 74-75: “The set of candidate viewpoints is Yk = Y0\y1:k, where Y0 is the initial set of available locations and y1:k = {y1,y2,...,yk} is the history of visited locations. The next location is chosen as the one which yields the largest utility. The active classification problem is defined as follows. Given observations Z1:k = {Z1,Z2,...,Zk}made from visited locations y1:k and the current belief Bk at stage k, choose the next location yk+1 from the set of available locations Yk by maximising the utility function”).
It would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate moving to a viewpoint as taught by Patten with the system of Robinson, Vasquez, and Ahmad because it would improve the system by allowing it to navigate a space as well as classify objects in that space (Patten Page 1). This motivation for the combination of Robinson, Vasquez, Ahmad, and Patten is supported by KSR exemplary rationale (G) Some teaching, suggestion, or motivation in the prior art that would have led one of ordinary skill to modify the prior art reference or to combine prior art reference teachings to arrive at the claimed invention and exemplary rational (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results. MPEP 2141 (III).
Regarding claim 15, Robinson does not explicitly disclose the system wherein the belief state represents belief over object locations in the structure of the occupancy octree.
However, Vasquez teaches the system wherein the belief state represents belief over object locations in the structure of the occupancy octree (Vasquez Page 6: “To represent the object bounding box, Wbox, we use a probabilistic occupancy map based on the octomap structure [5], which is an octree with probabilistic occupancy estimation. See Fig. 3. In this representation each voxel has associated a probability of being occupied. We use a probabilistic octree because it is able to deal with noise on the sensor readings. From now on we refer to a probabilistic occupancy map as octree”).
It would have been obvious to combinate Robinson and Vasquez for the same reasons used for claim 14 above.
Regarding claim 16, Robinson does not explicitly disclose the system wherein the occupancy octree represents a search region's occupancy.
However, Vasquez teaches the system wherein the occupancy octree represents a search region's occupancy (Vasquez Page 6: “To represent the object bounding box, Wbox, we use a probabilistic occupancy map based on the octomap structure [5], which is an octree with probabilistic occupancy estimation. See Fig. 3. In this representation each voxel has associated a probability of being occupied. We use a probabilistic octree because it is able to deal with noise on the sensor readings. From now on we refer to a probabilistic occupancy map as octree”).
It would have been obvious to combinate Robinson and Vasquez for the same reasons used for claim 14 above.
Response to Arguments
Rejections Under 35 U.S.C. § 101
Applicant argues on pages 5 and 6 of Applicant’s arguments and remarks that claim 1 should not be rejected under 101 because it integrates any abstract idea into a practical application. Examiner respectfully disagrees. Because all of the claim limitations are directed either to a mental process or insignificant data gathering, the claim does not integrate the abstract idea into a practical application. MPEP 2106.04(II)(A)(2) (“if there are no additional claim elements besides the judicial exception, or if the additional claim elements merely recite another judicial exception, that is insufficient to integrate the judicial exception into a practical application. See, e.g., RecogniCorp, LLC v. Nintendo Co., 855 F.3d 1322, 1327, 122 USPQ2d 1377 (Fed. Cir. 2017) ("Adding one abstract idea (math) to another abstract idea (encoding and decoding) does not render the claim non-abstract"); Genetic Techs. Ltd. v. Merial LLC, 818 F.3d 1369, 1376, 118 USPQ2d 1541, 1546 (Fed. Cir. 2016) (eligibility "cannot be furnished by the unpatentable law of nature (or natural phenomenon or abstract idea) itself.").”). Further, the specific improvement of the invention is not claimed in claim 1 or reflected in its limitations. As such, the 101 rejection of claim 1 is maintained.
Claim 14 is rejected for similar reasons as for claim. Claim 14 claims a system of generic computer components along with a mental process of moving the robot to a specific viewpoint. There is no application of that system to a practical application. As such, the 101 rejection of claim 14 is maintained.
In light of Applicant’s amendments and arguments regarding claim 11, the 101 rejections of claim 11-13 have been withdrawn.
Rejections Under 35 U.S.C. § 103
Applicant’s arguments with respect to claim(s) 1-16 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
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 AIDAN KEUP whose telephone number is (703)756-4578. The examiner can normally be reached Monday - Friday 8:00-4:00.
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, Emily Terrell can be reached at (571) 270-3717. 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.
/AIDAN KEUP/ Examiner, Art Unit 2666 /Molly Wilburn/Primary Examiner, Art Unit 2666
Prosecution Insights