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
The amendments to the claims, filed on 01/07/2026, have been entered and made of record.
Claims 1-20 are pending with claims 1, 2, and 10-12 being amended.
Response to Arguments
Arguments presented in the Remarks (“Remarks") filed on 01/07/2026 have been fully considered, but are rendered moot in view of the new ground(s) of rejection necessitated by amendment(s) initiated by the applicant(s).
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 of this title, 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.
Claims 1, 8, 10, 11, and 18 rejected under 35 U.S.C. 103 as being unpatentable over Ma et al. (“Ma”) [U.S Patent No. 10,832,439 B1] in view of Deo et al. (“Deo”) [U.S Patent Application Pub. No. 2022/0355825 A1] in further view of Marchetti-Bowick et al. (“Bowick”) [US 2021/0004012 A1]
Regarding claim 1, Ma meets the claim limitations as follows:
A method for vehicle motion forecasting, comprising [Fig. 1, 2; col. 3: the autonomous vehicle 102 … via a drive path 114]:
generating a lane graph structure (i.e. ‘Entity Location 120’) [Fig. 1: ‘120’, Fig. 2: ; ‘220’; col. 5, 7: ‘the entity location component 120 … functionality to identify a road segment and/or a lane of the road segment’; ‘to determine a lane’] according to a raw map data (i.e. ‘one or map(s) 118’) [Fig. 1, 2, 3: ‘118’; ‘202’; col. 3-5, 7, 8: ‘2D maps of a road network’ or ‘a 3D mesh’; ‘At operation 202, … receiving map information’];
establishing a plurality of occupancy flow graphs (i.e. ‘Graph Generation 124’) [Fig. 1: ‘124’; col. 6: ‘The graph generation component 124 … functionality to generate one or more directed graphs 126; Figs. 2-5: ‘224’; col. 8] which are homogeneous to data format of the lane graph structure according to trajectory data (i.e. ‘group vehicles according to lanes in which they are currently traveling and/or lanes in which they may travel in the future’) [Figs. 2-5; col. 5, ll. 43-67] of a plurality of vehicles in a plurality of consecutive frames (e.g. consecutive maps 304 and 318) [Figs. 2-5: ‘204’, ‘214’, ‘222’, ‘226’, ‘304’, ‘318’, …; ‘Generate updated directed graph based on the updated positions 322’; col. 9] and the lane graph structure [Figs. 2-5; col. 5];
establishing a plurality of temporal edges between the occupancy flow graphs (i.e. ‘Graph Generation 124’) [col. 5, 6, 8, 24: ‘The graph generation component 124 may also generate edges connecting the nodes to complete the directed graphs 126’; ‘updating a directed graph’; ‘Also in the directed graph 228, interactions or associations between the nodes are represented as edges 232’] according to the trajectory data of the vehicles in the consecutive frames, to construct a temporal occupancy flow graph [Figs. 1-5; col. 5, 6, 8, 24]; and
performing feature aggregation (e.g. ‘collectively, the additional vehicles 312’, ‘the movement (real or simulated) of entities within the road segments’ or ‘changed entity positions’) [col. 9] on the temporal occupancy flow graph to generating a plurality of updated node features [col. 5, 6, 8-10, 14, 24: ‘At operation 316, … At operation 322 … generating an updated directed graph’; ‘updating a directed graph in response to changed entity position’; ‘the updated directed graph 324 may be further updated to remove the edge extending between the vehicle node and the node 328’; ‘At operation 556, the process 500 can investigate edges between the determined node pairs to update the graph’], and generating a motion prediction (i.e. ‘a predicted position’) [col. 12, 24: ‘a prediction system 1024, a planning system 1026’; ‘the prediction system 1024 can determine features associated with the object’; ‘analyze features of objects to predict future actions of objects’] of an ego-vehicle according to the updated node features,
wherein the temporal edges are configured
Ma does not disclose explicitly the following claim limitations (emphasis added):
generating a motion prediction of an ego-vehicle according to the updated node features,
wherein the temporal edges are configured to connect a plurality of lane segments occupied by each of the vehicles in the occupancy flow graphs to each other, in order to express the trajectory data in the consecutive frames.
However in the same field of endeavor Deo discloses the deficient claim as follows:
generating a motion prediction (i.e. ‘to predict trajectories based on node encodings’) [Fig. 8, 14; para. 0138, 0148: ‘to learn a policy for sampled graph traversals based on a motion of a target vehicle as well as … at neighboring nodes’] of an ego-vehicle according to the updated node features,
wherein the temporal edges are configured to connect a plurality of lane segments occupied by each of the vehicles in the occupancy flow graphs to each other, in order to express the trajectory data in the consecutive frames.
Ma and Deo are combinable because they are from the same field of planning systems for autonomous vehicles [Ma and Deo: Background Section.].
It would have been obvious to one with ordinary skill in the art before the effective filling date of the claimed invention to combine teachings of Ma and Deo as motivation to predict vehicle trajectories for safe and efficient navigation through complex traffic scenes [Deo: Background].
Deo does not disclose explicitly the following claim limitations (emphasis added):
wherein the temporal edges are configured to connect a plurality of lane segments occupied by each of the vehicles in the occupancy flow graphs to each other, in order to express the trajectory data in the consecutive frames.
However in the same field of endeavor Bowick discloses the deficient claim as follows: wherein the temporal edges are configured to connect a plurality of lane segments occupied by each of the vehicles in the occupancy flow graphs to each other [Fig. 6, 9; para. 0024, 0038, 0039, 0084: ‘a set of candidate paths can be represented using a lane graph where nodes represent lane segments and nodes represent connections between lane segments’; ‘a lane graph can be generated for each of a plurality of objects in a scene’; ‘Each edge can have a type that identifies the type of connection between the lane segments. For example, the edge may include a type that identifies whether it is a predecessor or successor to another segment, whether it is adjacent to another segment, whether it is opposing to another segment, and/or whether it is conflicting with another segment’], in order to express the trajectory data in the consecutive frames.
Ma, Deo and Bowick are combinable because they are from the same field of planning systems for autonomous vehicles [Ma and Deo: Background Section.].
It would have been obvious to one with ordinary skill in the art before the effective filling date of the claimed invention to combine teachings of Ma, Deo and Bowick as motivation to predict vehicle trajectories for safe and efficient navigation through complex traffic scenes [Deo: Background; Bowick: para. 0024-0026: ‘improved systems and methods for predicting the future behavior … of objects’].
Regarding claim 8, Ma meets the claim limitations as follows:
The method of claim 1, wherein step of generating the motion prediction of the ego-vehicle comprises: inputting the updated node features and a past trajectory of the ego-vehicle in the consecutive frames to a downstream model (i.e. ‘prediction’ and ‘planning’ model) [Fig. 10: ‘the prediction system 1024’, ‘the planning system 1026’] , as such the downstream model generates the motion prediction of the ego-vehicle (i.e. ‘a predicted position’) [col. 2, 12, 22, 24: ‘a prediction system 1024, a planning system 1026’; ‘the prediction system 1024 can determine features associated with the object’; ‘analyze features of objects to predict future actions of objects’].
Ma does not disclose explicitly the following claim limitations (emphasis added):
generating a motion prediction of an ego-vehicle according to the updated node features.
However in the same field of endeavor Deo discloses the deficient claim as follows:
generating a motion prediction (i.e. ‘to predict trajectories based on node encodings’) [Fig. 8, 14; para. 0138, 0148: ‘to learn a policy for sampled graph traversals based on a motion of a target vehicle as well as … at neighboring nodes’] of an ego-vehicle according to the updated node features.
Ma and Deo are combinable because they are from the same field of planning systems for autonomous vehicles [Ma and Deo: Background Section.].
It would have been obvious to one with ordinary skill in the art before the effective filling date of the claimed invention to combine teachings of Ma and Deo as motivation to predict vehicle trajectories for safe and efficient navigation through complex traffic scenes [Deo: Background].
Regarding claim 10, Ma meets the claim limitations as follows:
A method for vehicle motion forecasting, comprising [Fig. 1, 2; col. 3: the autonomous vehicle 102 … via a drive path 114]:
generating a lane graph structure according to a raw map data [See rejection of claim 1 limitation: ‘generating a lane graph structure’];
mapping a plurality of first bounding boxes of a plurality of vehicles [Fig. 6: ‘610’; ‘AABBs’; ‘612’; col. 7, ll. 45-65: ‘the point may be a point on a bounding box defining the vehicle 216’; col. 15, ll. 43-65; col. 16, 17] in a first frame onto the lane graph structure to generate a first occupancy flow graph (i.e. ‘Graph Generation 124’) [Fig. 1: ‘124’; col. 6: ‘The graph generation component 124 … functionality to generate one or more directed graphs 126; Figs. 2-6: ‘224’; col. 8];
mapping a plurality of second bounding boxes of the vehicles [Fig. 6: ‘610’; ‘AABBs’; ‘612’; col. 7, ll. 45-65: ‘the point may be a point on a bounding box defining the vehicle 216’; col. 15, ll. 43-65; col. 16, 17] in a second frame onto the lane graph structure to generate a second occupancy flow graph (i.e. ‘Graph Generation 124’) [Fig. 1: ‘124’; col. 6: ‘The graph generation component 124 … functionality to generate one or more directed graphs 126; Figs. 2-6: ‘224’; col. 8], and
wherein the first frame and the second frame are consecutive frames [Figs. 2-5 show consecutive maps (i.e. frames); col. 5, 6, 8, 24: ‘The graph generation component 124 may also generate edges connecting the nodes to complete the directed graphs 126’; ‘updating a directed graph’; ‘Also in the directed graph 228, interactions or associations between the nodes are represented as edges 232’];
establishing temporal edges between the first occupancy flow graph and the second occupancy flow graph to construct a temporal occupancy flow graph [See rejection of claim 1 limitation: ‘establishing a plurality of temporal edges’]; and
performing feature aggregation on the temporal occupancy flow graph to generate a plurality of updated node features [See rejection of claim 1 limitation: ‘performing feature aggregation’], and generating a motion prediction of an ego-vehicle according to the updated node features [See rejection of claim 1 limitation: ‘generating a motion prediction’],
wherein the temporal edges are configured to connect a plurality of lane segments occupied by each of the vehicles in the occupancy flow graphs to each other, in order to express the trajectory data in the consecutive frames.
Ma and Deo are combinable because they are from the same field of planning systems for autonomous vehicles [Ma and Deo: Background Section.].
It would have been obvious to one with ordinary skill in the art before the effective filling date of the claimed invention to combine teachings of Ma and Deo as motivation to predict vehicle trajectories for safe and efficient navigation through complex traffic scenes [Deo: Background].
Deo does not disclose explicitly the following claim limitations (emphasis added):
wherein the temporal edges are configured to connect a plurality of lane segments occupied by each of the vehicles in the occupancy flow graphs to each other, in order to express the trajectory data in the consecutive frames.
However in the same field of endeavor Bowick discloses the deficient claim as follows: wherein the temporal edges are configured to connect a plurality of lane segments occupied by each of the vehicles in the occupancy flow graphs to each other [Fig. 6, 9; para. 0024, 0038, 0039, 0084: ‘a set of candidate paths can be represented using a lane graph where nodes represent lane segments and nodes represent connections between lane segments’; ‘a lane graph can be generated for each of a plurality of objects in a scene’; ‘Each edge can have a type that identifies the type of connection between the lane segments. For example, the edge may include a type that identifies whether it is a predecessor or successor to another segment, whether it is adjacent to another segment, whether it is opposing to another segment, and/or whether it is conflicting with another segment’], in order to express the trajectory data in the consecutive frames.
Ma, Deo and Bowick are combinable because they are from the same field of planning systems for autonomous vehicles [Ma and Deo: Background Section.].
It would have been obvious to one with ordinary skill in the art before the effective filling date of the claimed invention to combine teachings of Ma, Deo and Bowick as motivation to predict vehicle trajectories for safe and efficient navigation through complex traffic scenes [Deo: Background; Bowick: para. 0024-0026: ‘improved systems and methods for predicting the future behavior … of objects’].
Regarding claim 11, all claim limitations are set forth as claim 1 in the system form and rejected as per discussion for claim 1. Both Ma and Deo discloses “a memory” and “a processing circuit” [Ma: col. 19, ll. 52-65; col. 27, ll. 50-67; Deo: para. 0057, 0061-0062].
Regarding claim 18, all claim limitations are set forth as claim 8 in the system form and rejected as per discussion for claim 8.
Claims 2-4, 6-7, 12-14 and 16-17 rejected under 35 U.S.C. 103 as being unpatentable over Ma in view of Deo in further view of Bowick in further view of Laine et al. (“Laine”) [U.S Patent Application Pub. No. 2016/0071312 A1]
Regarding claim 2, Ma meets the claim limitations as follows:
The method of claim 1, wherein step of establishing each of the occupancy flow graphs comprises:
receiving a plurality of bounding boxes (i.e. ‘612’) of the vehicles in a frame [Fig. 6: ‘AABBs’; col. 7, ll. 45-65: ‘the point may be a point on a bounding box defining the vehicle 216’; col. 15, ll. 43-65; col. 16, 17];
inheriting (i.e. KD tree structure) [col. 5] a plurality of lane segment features of the lane segments and a plurality of geometric edges from the lane graph structure [Figs. 2-5: ‘226’; ‘304’; ‘328’; ‘404’; ‘558’; Fig. 6: ‘bounding box’];
computing a plurality of occupant vehicle features of the lane segments according to the bounding boxes and speed information of the vehicles in the frame [col. 5, 6, 8, 24: ‘The graph generation component 124 may also generate edges connecting the nodes to complete the directed graphs 126’; ‘updating a directed graph’; ‘Also in the directed graph 228, interactions or associations between the nodes are represented as edges 232’]; and
establishing a plurality of vehicle interaction edges by connecting a portion of the lane segments occupied by the bounding boxes [Fig. 6: ‘AABBs’; col. 7, ll. 45-65: ‘the point may be a point on a bounding box defining the vehicle 216’; col. 15, ll. 43-65; col. 16, 17], correspondingly [col. 5, 6, 8, 24: ‘The graph generation component 124 may also generate edges connecting the nodes to complete the directed graphs 126’; ‘updating a directed graph’; ‘Also in the directed graph 228, interactions or associations between the nodes are represented as edges 232’].
Ma does not disclose explicitly the following claim limitations (emphasis added):
inheriting a plurality of lane segment features of the lane segments and a plurality of geometric edges from the lane graph structure.
However in the same field of endeavor Laine discloses the deficient claim as follows:
inheriting (i.e. k-d tree or bounding volume hierarchy) [para. 0004, 0137-0139, 0141-0143: ‘the child node may specify which values that define the six planes of the AABB are new, … and which values are inherited from the AABB of the parent node’] a plurality of lane segment features of the lane segments and a plurality of geometric edges from the lane graph structure.
Ma and Laine are combinable because they are from the same field of “axis-aligned bounding boxes (AABB)’.
It would have been obvious to one with ordinary skill in the art before the effective filling date of the claimed invention to combine teachings of Ma and Laine as motivation to include inheritance to a child node in the AABB tree as a solution to the problems cited in Laine [para. 0005] .
Regarding claim 3, Ma meets the claim limitations as follows:
The method of claim 2, wherein the lane segment features include a start point, an end point and a centroid of each of the lane segments [Figs. 2-5, 9 disclose start and end point and ‘segment-centric coordinate system’; col. 5, 7, 18-20: ‘a center of the corresponding one of the vehicles 216’], and wherein the geometric edges are connection between adjacent two of the lane segments based on drivable path [Figs. 2-5, 9 disclose start point, end point, edges and ‘segment-centric coordinate system’; col. 6, 10, 24: use edges to connect related nodes].
Regarding claim 4, Ma meets the claim limitations as follows:
The method of claim 2, wherein step of computing the occupant vehicle features comprising: computing a vehicle occupancy value of each of the lane segments [col. 13: ‘each of the edges 532 represents a pair of values’; col. 30: ‘a path of travel of the autonomous vehicle in the road segment …’] according to the bounding boxes; and computing an occupancy flow vector of each of the lane segments according to the speed information of the vehicles in the frame [col. 15, 23: ‘each junction 608 may have its own characteristics, …,including, but not limited to, information about a direction of travel, .., speed limits’; ‘a velocity of the object’].
Ma does not disclose explicitly the following claim limitations (emphasis added):
computing a vehicle occupancy value of each of the lane segments according to the bounding boxes; and computing an occupancy flow vector of each of the lane segments according to the speed information of the vehicles in the frame.
However in the same field of endeavor Deo discloses the deficient claim as follows:
computing a vehicle occupancy value of each of the lane segments according to the bounding boxes [para. 0080-0081: ‘a plurality of feature values referred to as F1, F2, … FN]; and computing an occupancy flow vector of each of the lane segments according to the speed information [para. 0033, 0069: ‘routes 106 may include more precise actions or states such as, for example, specific target lanes or precise locations within the lane areas and targeted speed at those positions’] of the vehicles in the frame.
Ma and Deo are combinable because they are from the same field of planning systems for autonomous vehicles [Ma and Deo: Background Section.].
It would have been obvious to one with ordinary skill in the art before the effective filling date of the claimed invention to combine teachings of Ma and Deo as motivation to predict vehicle trajectories for safe and efficient navigation through complex traffic scenes [Deo: Background].
Regarding claim 6, Ma meets the claim limitations as follows:
The method of claim 2, wherein step of constructing the temporal occupancy flow graph comprising: establishing the temporal edges by connecting a plurality of occupied lane segments occupied by the same vehicle in two consecutive frames [Figs. 2-6; Fig. 3: two temporal frames 304 and 318 and edges connecting lane segments, e.g. 314 (1), 314(2), … 314(7); col. 6: ‘The graph generation component 124 may also generate edges connecting the nodes’], correspondingly.
Regarding claim 7, Ma meets the claim limitations as follows:
The method of claim 2, wherein step of performing the feature aggregation on the temporal occupancy flow graph comprises: extracting, from the temporal occupancy flow graph, interaction information between the vehicles and interaction information between the vehicles [Fig. 3: two frames ‘304’ and ‘318’ show the vehicle 312(2) making a left turn and at different positions of 312(2) in the two frames’] and the lane segments at the same time according to the geometric edges and vehicle interaction edges of each of the occupancy flow graphs [Figs. 2-6; Fig. 3: two temporal frames 304 and 318 and edges connecting lane segments, e.g. 314 (1), 314(2), … 314(7); col. 6: ‘The graph generation component 124 may also generate edges connecting the nodes’].
Regarding claim 12, all claim limitations are set forth as claim 2 in the system form and rejected as per discussion for claim 2.
Regarding claim 13, all claim limitations are set forth as claim 3 in the system form and rejected as per discussion for claim 3.
Regarding claim 14, all claim limitations are set forth as claim 4 in the system form and rejected as per discussion for claim 4.
Regarding claim 16, all claim limitations are set forth as claim 6 in the system form and rejected as per discussion for claim 6.
Regarding claim 17, all claim limitations are set forth as claim 7 in the system form and rejected as per discussion for claim 7.
Claims 9, 19 and 20 rejected under 35 U.S.C. 103 as being unpatentable over Ma in view of Deo in further view of Bowick in further view of Hopecraft et al. (“Hopecraft”) [U.S Patent Application Pub. No. 2016/0205509 A1]
Regarding claim 9, Ma in view of Deo and Bowick meets the claim limitations as follows:
The method of claim 1, further comprising: embedding traffic light status (e.g. ‘the left-hand turn’) [col. 5, 6, 8, 26: ‘the communication connection(s) 1010 can allow the vehicle to communicate with other nearby computing device(s) (e.g., other nearby vehicles, traffic signals, etc.)’. Note: Deo: para. 0044: ‘one or more images associated with one or more traffic lights’] and lane information [See claim 1 rejection of limitations: ‘establishing a plurality of occupancy flow graphs’] to the temporal occupancy flow graph to establish a temporal occupancy flow graph with traffic information [See claim 1 rejection of limitations: ‘establishing a plurality of temporal edges’]; and performing the feature aggregation on the temporal occupancy flow graph with the traffic information to generate the updated node features, and generating the motion prediction of the ego-vehicle according to the updated node features [See claim 1 rejection of limitations: ‘performing feature aggregation’ and ‘generating a motion prediction’].
Ma does not disclose explicitly the following claim limitations (emphasis added):
embedding traffic light status and lane information to the temporal occupancy flow graph to establish a temporal occupancy flow graph with traffic information.
However in the same field of endeavor Hopecraft discloses the deficient claim as follows:
embedding traffic light status [para. 0034: ‘further information may be appended to the graph data structure, such as the road signs and markings and traffic lights’] and lane information to the temporal occupancy flow graph to establish a temporal occupancy flow graph with traffic information.
Ma, Deo, Bowick and Hopecraft are combinable because they are from the same field of planning systems for autonomous vehicles.
It would have been obvious to one with ordinary skill in the art before the effective filling date of the claimed invention to combine teachings of Ma, Deo, Bowick and Hopecraft as motivation to append further information to the graph data structure for determining the location of an object and tracking the object [Hopecraft: para. 0001].
Regarding claim 19, all claim limitations are set forth as claim 9 in the system form and rejected as per discussion for claim 9.
Regarding claim 20, all claim limitations are set forth as claim 11 in the system form and rejected as per discussion for claim 11.
Allowable Subject Matter
Regarding claim 5, it is objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Regarding claim 15, it is objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Conclusion
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 extension fee 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 PETER D LE whose telephone number is (571)270-5382. The examiner can normally be reached on Monday - Alternate Friday: 10AM-6:30PM.
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, SATH PERUNGAVOOR can be reached on 571-272-7455. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/PETER D LE/
Primary Examiner, Art Unit 2488