公开(公告)号：US12271204B1

公开(公告)日：2025-04-08

申请号：US17081203

申请日：2020-10-27

Applicant: Zoox, Inc.

Inventor： Gowtham Garimella ,  Marin Kobilarov ,  Kai Zhenyu Wang

IPC: G06N20/00 ,  G05D1/00 ,  G06V20/58

Abstract: Techniques are discussed for predicting an occupancy of visible region of an environment. For instance, a vehicle may generate sensor data representing an environment. The vehicle may then analyze the sensor data to determine an occluded region of the environment a visible region of the environment. Additionally, the vehicle may determine at least one prediction probability associated with occupancy of the visible region over a future period of time. In some instances, the vehicle determines the at least one prediction probability by inputting data representing at least the occluded region and the visible region into a machine learned model and receiving the at least one prediction probability from the machine learned model. Using the at least one prediction probability, the vehicle may then determine and perform one or more actions.

公开(公告)号：US12084087B2

公开(公告)日：2024-09-10

申请号：US17535418

申请日：2021-11-24

Applicant: Zoox, Inc.

Inventor： Gowtham Garimella ,  Marin Kobilarov ,  Andres Guillermo Morales Morales ,  Ethan Miller Pronovost ,  Kai Zhenyu Wang ,  Xiaosi Zeng

IPC: B60W60/00 ,  G06N3/02

CPC classification number: B60W60/0027 ,  G06N3/02 ,  B60W2554/402 ,  B60W2554/4041 ,  B60W2554/4042 ,  B60W2554/4043 ,  B60W2554/4045 ,  B60W2554/4046 ,  B60W2555/60 ,  B60W2556/40

Abstract: Techniques for determining unified futures of objects in an environment are discussed herein. Techniques may include determining a first feature associated with an object in an environment and a second feature associated with the environment and based on a position of the object in the environment, updating a graph neural network (GNN) to encode the first feature and second feature into a graph node representing the object and encode relative positions of additional objects in the environment into one or more edges attached to the node. The GNN may be decoded to determine a distribution of predicted positions for the object in the future that meet a criterion, allowing for more efficient sampling. A predicted position of the object in the future may be determined by sampling from the distribution.

公开(公告)号：US12080044B2

公开(公告)日：2024-09-03

申请号：US17535396

申请日：2021-11-24

Applicant: Zoox, Inc.

Inventor： Gowtham Garimella ,  Marin Kobilarov ,  Andres Guillermo Morales Morales ,  Ethan Miller Pronovost ,  Kai Zhenyu Wang ,  Xiaosi Zeng

IPC: G06V10/44 ,  B60W60/00 ,  G06F16/901 ,  G06V10/422 ,  G06V10/764

CPC classification number: G06V10/454 ,  B60W60/0027 ,  G06F16/9024 ,  G06V10/422 ,  G06V10/764

Abstract: Techniques for determining unified futures of objects in an environment are discussed herein. Techniques may include determining a first feature associated with an object in an environment and a second feature associated with the environment and based on a position of the object in the environment, updating a graph neural network (GNN) to encode the first feature and second feature into a graph node representing the object and encode relative positions of additional objects in the environment into one or more edges attached to the node. The GNN may be decoded to determine a distribution of predicted positions for the object in the future. A predicted position of the object at a subsequent timestep may be determined by sampling from the distribution of predicted positions according to various sampling strategies. Alternatively, the predicted position of the object may be overwritten using a candidate position of the object.

公开(公告)号：US11891088B1

公开(公告)日：2024-02-06

申请号：US17347088

申请日：2021-06-14

Applicant: Zoox, Inc.

Inventor： Marin Kobilarov ,  Jefferson Bradfield Packer ,  Gowtham Garimella ,  Andreas Pasternak ,  Yiteng Zhang ,  Ruikun Yu

IPC: B60W60/00 ,  G06N20/00 ,  G07C5/08 ,  G07C5/00 ,  B60W50/00

CPC classification number: B60W60/0015 ,  G06N20/00 ,  G07C5/008 ,  G07C5/0808 ,  B60W2050/0075 ,  B60W2554/404 ,  B60W2556/45

Abstract: A reward determined as part of a machine learning technique, such as reinforcement learning, may be used to control an adversarial agent in a simulation such that a component for controlling motion of the adversarial agent is trained to reduce the reward. Training the adversarial agent component may be subject to one or more constraints and/or may be balanced against one or more additional goals. Additionally or alternatively, the reward may be used to alter scenario data so that the scenario data reduces the reward, allowing the discovery of difficult scenarios and/or prospective events.

公开(公告)号：US20240001958A1

公开(公告)日：2024-01-04

申请号：US17854849

申请日：2022-06-30

Applicant: Zoox, Inc.

Inventor： Gowtham Garimella ,  Gary Linscott ,  Ethan Miller Pronovost

IPC: B60W60/00 ,  B60W30/095 ,  B60W40/04 ,  B60W50/00

CPC classification number: B60W60/0011 ,  B60W60/0015 ,  B60W60/00274 ,  B60W2556/40 ,  B60W40/04 ,  B60W50/0097 ,  B60W2554/4041 ,  B60W30/0956

Abstract: Techniques for improving operational decisions of an autonomous vehicle are discussed herein. In some cases, a system may generate reference graphs associated with a route of the autonomous vehicle. Such reference graphs can comprise precomputed feature vectors based on grid regions and/or lane segments. The feature vectors are usable to determine scene context data associated with static objects to reduce computational expenses and compute time.

公开(公告)号：US20230162470A1

公开(公告)日：2023-05-25

申请号：US17535396

申请日：2021-11-24

Applicant: Zoox, Inc.

Inventor： Gowtham Garimella ,  Marin Kobilarov ,  Andres Guillermo Morales Morales ,  Ethan Miller Pronovost ,  Kai Zhenyu Wang ,  Xiaosi Zeng

IPC: G06V10/44 ,  G06V10/422 ,  B60W60/00 ,  G06F16/901 ,  G06V10/764

CPC classification number: G06V10/454 ,  G06V10/422 ,  B60W60/0027 ,  G06F16/9024 ,  G06V10/764

Abstract: Techniques for determining unified futures of objects in an environment are discussed herein. Techniques may include determining a first feature associated with an object in an environment and a second feature associated with the environment and based on a position of the object in the environment, updating a graph neural network (GNN) to encode the first feature and second feature into a graph node representing the object and encode relative positions of additional objects in the environment into one or more edges attached to the node. The GNN may be decoded to determine a distribution of predicted positions for the object in the future. A predicted position of the object at a subsequent timestep may be determined by sampling from the distribution of predicted positions according to various sampling strategies. Alternatively, the predicted position of the object may be overwritten using a candidate position of the object.

公开(公告)号：US20250162616A1

公开(公告)日：2025-05-22

申请号：US18516618

申请日：2023-11-21

Applicant: Zoox, Inc.

Inventor： Gregory Michael Woelki ,  Xiaosi Zeng ,  Gowtham Garimella ,  Samir Parikh ,  Ethan Miller Pronovost

IPC: B60W60/00 ,  B60W50/06

Abstract: A machine-learned architecture may predict a set of spatially-diverse paths that an object may take in the future. The paths generated by this architecture may be time-invariant (e.g., not identifying a time at which the object may occupy a position along one of these paths) but can be used by a second machine-learned model to predict progress in time along these paths. This segregation of the spatial paths and progress in time along the paths improves the accuracy of the ultimate prediction and better captures rare object behavior.

公开(公告)号：US12269462B1

公开(公告)日：2025-04-08

申请号：US16820378

申请日：2020-03-16

Applicant: Zoox, Inc.

Inventor： Gowtham Garimella ,  Marin Kobilarov ,  Kai Zhenyu Wang

IPC: B60W30/00 ,  B60W30/06 ,  B62D15/02 ,  G06N3/08

Abstract: Techniques relating to determining regions based on intents of objects are described. In an example, a computing device onboard a first vehicle can receive sensor data associated with an environment of the first vehicle. The computing device can determine, based on the sensor data, a region associated with a second vehicle proximate the first vehicle that is to be occupied by the second vehicle while the vehicle performs a maneuver. Further, the computing device can determine an instruction for controlling the first vehicle based at least in part on the region.

公开(公告)号：US11858514B2

公开(公告)日：2024-01-02

申请号：US17218051

申请日：2021-03-30

Applicant: Zoox, Inc.

Inventor： Gerrit Bagschik ,  Andrew Scott Crego ,  Gowtham Garimella ,  Michael Haggblade ,  Andraz Kavalar ,  Kai Zhenyu Wang

IPC: B60W30/18 ,  G06N3/088 ,  B60W50/14 ,  G06N3/045 ,  B60W30/095 ,  B60W60/00 ,  G06N3/08 ,  G06V20/58

CPC classification number: B60W30/18009 ,  B60W30/0956 ,  B60W50/14 ,  B60W60/0011 ,  G06N3/045 ,  G06N3/08 ,  G06N3/088 ,  G06V20/58 ,  B60W2556/10

Abstract: Techniques for top-down scene discrimination are discussed. A system receives scene data associated with an environment proximate a vehicle. The scene data is input to a convolutional neural network (CNN) discriminator trained using a generator and a classification of the output of the CNN discriminator. The CNN discriminator generates an indication of whether the scene data is a generated scene or a captured scene. If the scene data is data generated scene, the system generates a caution notification indicating that a current environmental situation is different from any previous situations. Additionally, the caution notification is communicated to at least one of a vehicle system or a remote vehicle monitoring system.

公开(公告)号：US11810365B1

公开(公告)日：2023-11-07

申请号：US17122998

申请日：2020-12-15

Applicant: Zoox, Inc.

Inventor： Andrew Scott Crego ,  Gowtham Garimella ,  Mahsa Ghafarianzadeh ,  Rasmus Fonseca ,  Muhammad Farooq Rama ,  Kai Zhenyu Wang

IPC: G06V20/58 ,  B60W60/00 ,  G06F18/213 ,  G06F18/214 ,  G06F18/2415

CPC classification number: G06V20/58 ,  B60W60/001 ,  G06F18/213 ,  G06F18/214 ,  G06F18/2415 ,  B60W2554/4049

Abstract: Techniques for modeling the probability distribution of errors in perception systems are discussed herein. For example, techniques may include modeling error distribution for attributes such as position, size, pose, and velocity of objects detected in an environment, and training a mixture model to output specific error probability distributions based on input features such as object classification, distance to the object, and occlusion. The output of the trained model may be used to control the operation of a vehicle in an environment, generate simulations, perform collision probability analyses, and to mine log data to detect collision risks.