公开(公告)号：US20240203130A1

公开(公告)日：2024-06-20

申请号：US18065711

申请日：2022-12-14

Applicant: Kodiak Robotics, Inc.

Inventor： Suchir Gupta ,  Collin C. Otis ,  Cole M. Miles ,  Philip C. Du Toit ,  Andreas Wendel

IPC: G06V20/58 ,  G06N20/00

CPC classification number: G06V20/58 ,  G06N20/00

Abstract: This disclosure provides methods and systems for detecting and tracking objects in an environment of an autonomous vehicle. The method may include: receiving sensor data from at least one sensor of the autonomous vehicle, the sensor data representative of one or more portions of an object in the environment of the autonomous vehicle; determining a highest confidence portion of the object, wherein the highest confidence portion of the object comprises a portion of the object that is observed and estimated with highest accuracy and confidence; determining features of the highest confidence portion of the object; training a machine learning model based at least in part on the features of the highest confidence portion of the object and an error metric that measures difference between the highest confidence portion of the object and a corresponding portion of the object in a ground truth; and detecting or tracking one or more objects in the environment of the autonomous vehicle using the trained machine learning model.

公开(公告)号：US20240190463A1

公开(公告)日：2024-06-13

申请号：US18179097

申请日：2023-03-06

Applicant: Kodiak Robotics, Inc.

Inventor： Derek J. Phillips ,  Collin C. Otis ,  Andreas Wendel ,  Jackson P. Rusch

IPC: B60W60/00 ,  B60W30/09 ,  B60W40/02

CPC classification number: B60W60/001 ,  B60W30/09 ,  B60W40/02 ,  B60W2420/00 ,  B60W2554/00 ,  B60W2556/25

Abstract: This disclosure provides systems and methods for path planning by a planner of an autonomous vehicle. The method may include receiving by the planner perception data from a perception module, wherein the perception data comprises tracking or predicted object data associated with objects or obstacles in an environment of the autonomous vehicle, and wherein the tracking or predicted object data are determined based on high recall detection data and high precision detection data, generating by the planner a trajectory for controlling the autonomous vehicle based on the perception data received from the perception module, and transmitting to a controller of the autonomous vehicle the trajectory such that the autonomous vehicle is navigated by the controller to a destination.

公开(公告)号：US20240185612A1

公开(公告)日：2024-06-06

申请号：US18062228

申请日：2022-12-06

Applicant: Kodiak Robotics, Inc.

Inventor： Ryan T. Lee ,  Derek J. Phillips ,  Collin C. Otis

IPC: G06V20/58 ,  G06V10/762 ,  G06V10/82

CPC classification number: G06V20/58 ,  G06V10/762 ,  G06V10/82

Abstract: This disclosure provides methods and systems for dynamically detecting and tracking objects in an environment of an autonomous vehicle. In some embodiments, the method comprises: receiving image data from sensors of the autonomous vehicle, the image data comprising a plurality of images representative of objects in a field of view of the autonomous vehicle; detecting the objects in the plurality of images through an object detector; generating image embeddings for the objects detected in the plurality of images; determining similarity scores of the image embeddings of the objects that are detected in images received from two or more different sensors; identifying the objects that are detected in the images received from the two or more different sensors as a candidate object for tracking, if the objects have a similarity score of the image embeddings equal to or greater than a threshold value; and initializing a track for the candidate object.

公开(公告)号：US20240304001A1

公开(公告)日：2024-09-12

申请号：US18179075

申请日：2023-03-06

Applicant: Kodiak Robotics, Inc.

Inventor： Kenar Vyas ,  Sean Roelofs ,  Collin C. Otis ,  Philip C. Du Toit

IPC: G06V20/56 ,  G01S17/86 ,  G01S17/894 ,  G01S17/931 ,  G06V10/82

CPC classification number: G06V20/588 ,  G01S17/86 ,  G01S17/894 ,  G01S17/931 ,  G06V10/82

Abstract: Systems and methods for detecting a portion of an environment of a vehicle are provided. The method may comprise generating, using one or more sensors coupled to a vehicle, environment data from an environment of the vehicle, wherein the environmental data comprises one or more of the following: ground LiDAR data from the environment; camera data from the environment; and path data corresponding to a change in position of one or more other vehicles within the environment. The method may comprise inputting the environmental data into a machine learning model trained to generate a heatmap, and, using a processor, based on the environmental data, determining a portion of the environment, wherein the portion of the environment comprises an area having a likelihood, greater than a minimum threshold, of being adjacent to one or more pavement markings, and generating the heatmap, wherein the heatmap corresponds to the portion of the environment.

公开(公告)号：US20240199065A1

公开(公告)日：2024-06-20

申请号：US18147190

申请日：2022-12-28

Applicant: Kodiak Robotics, Inc.

Inventor： Suchir Gupta ,  Collin C. Otis ,  Cole M. Miles ,  Philip C. Du Toit ,  Andreas Wendel

IPC: B60W60/00 ,  G06N3/0464 ,  G06V20/58

CPC classification number: B60W60/001 ,  G06N3/0464 ,  G06V20/58 ,  B60W2554/4041

Abstract: This disclosure provides methods and systems for generating a training set for a neural network configured to generate candidate trajectories for an autonomous vehicle, comprising: receiving a set of sensor data representative of one or more portions of a plurality of objects in the environment of the autonomous vehicle; for each object, calculating a representative box enclosing the object, the representative box having portions comprising corners, edges, and planes; for each representative box, calculating at least one vector into the representative box from a position on the autonomous vehicle; for each vector, calculating a first and second corner position of the representative box, an edge of the representative box, and a plane of the representative box; determining the highest confidence corners, edge, and plane of each representative based on calculation from the at least one vector; and generating a training set including the highest confidence corners, edge, and plane of each representative box.

公开(公告)号：US20240190467A1

公开(公告)日：2024-06-13

申请号：US18065421

申请日：2022-12-13

Applicant: Kodiak Robotics, Inc.

Inventor： Derek J. Phillips ,  Collin C. Otis ,  Andreas Wendel ,  Jackson P. Rusch

IPC: B60W60/00 ,  B60W30/09 ,  B60W30/095 ,  B60W40/04

CPC classification number: B60W60/0011 ,  B60W30/09 ,  B60W30/0956 ,  B60W40/04 ,  B60W60/0015 ,  B60W2420/42 ,  B60W2420/52 ,  B60W2420/54 ,  B60W2554/4044 ,  B60W2554/4045 ,  B60W2556/25

Abstract: This disclosure provides systems and methods for controlling a vehicle. The method comprises receiving data from a set of sensors, wherein the data represents objects or obstacles in an environment of the autonomous vehicle; determining attributes of each of the objects or obstacles based on the received data from the set of sensors; integrating the attributes of the each of the objects or obstacles; identifying objects or obstacles based on the integrated attributes; determining a candidate trajectory for the autonomous vehicle to avoid the objects or obstacles; and controlling the autonomous vehicle according to the candidate trajectory.

公开(公告)号：US20240239368A1

公开(公告)日：2024-07-18

申请号：US18155378

申请日：2023-01-17

Applicant: Kodiak Robotics, Inc.

Inventor： Philip C. Du Toit ,  Akshay Khatri ,  Andreas Wendel ,  Zachais Vawter ,  Collin C. Otis ,  Ryan T. Lee

IPC: B60W60/00 ,  G06V10/26 ,  G06V10/34 ,  G06V20/56

CPC classification number: B60W60/001 ,  G06V10/26 ,  G06V10/34 ,  G06V20/588 ,  B60W2552/00

Abstract: This disclosure provides methods and systems for dynamically creating a trajectory for navigating a vehicle. The method may include receiving sensor data from at least one sensor of the autonomous vehicle, the sensor data representative of a driving surface in a field of view of the autonomous vehicle; segmenting a portion of the driving surface in the field of view of the autonomous vehicle by determining nominal path based at least in part on the image data; assigning a plurality of nodes to at least a portion of the nominal path; associating the plurality of the nodes assigned to the nominal path with a line to generate at least one segmentation polyline; determining updated nominal path by fitting the each of the plurality of segmentation lines to the nominal path; generating a trajectory based on the updated nominal path; and navigating the autonomous vehicle according to the trajectory.

公开(公告)号：US20240190470A1

公开(公告)日：2024-06-13

申请号：US18065417

申请日：2022-12-13

Applicant: Kodiak Robotics, Inc.

Inventor： Derek J. Phillips ,  Collin C. Otis ,  Andreas Wendel ,  Jackson P. Rusch

IPC: B60W60/00 ,  G06T7/20 ,  G06V10/764 ,  G06V10/80 ,  G06V20/58

CPC classification number: B60W60/0015 ,  G06T7/20 ,  G06V10/764 ,  G06V10/803 ,  G06V20/58 ,  B60W2420/42 ,  B60W2420/52 ,  B60W2420/54 ,  B60W2420/62 ,  G06T2207/30241 ,  G06T2207/30261 ,  G06V2201/08

Abstract: This disclosure provides systems and methods for controlling a vehicle based on a combination of high precision detection and high recall detection. The disclosed systems and methods can efficiently generate trajectories by reducing duplicate detection or duplicate calculation of objects or obstacles of common and known object types and objects or obstacles without class identification.

公开(公告)号：US12258041B2

公开(公告)日：2025-03-25

申请号：US18065419

申请日：2022-12-13

Applicant: Kodiak Robotics, Inc.

Inventor： Derek J. Phillips ,  Collin C. Otis ,  Andreas Wendel ,  Jackson P. Rusch

IPC: B60W60/00 ,  G01S13/86

Abstract: This disclosure provides systems and methods for controlling a vehicle. The method comprises receiving data from a set of sensors, wherein the data represents objects or obstacles in an environment of the autonomous vehicle; identifying objects or obstacles from the received data; determining multiple sets of attributes of the objects or obstacles, wherein each set of attributes of the objects or obstacles are determined based on data received by an individual sensor; determining a candidate trajectory for the autonomous vehicle based on the multiple sets of attributes of the objects or obstacles; and controlling the autonomous vehicle according to the candidate trajectory.

公开(公告)号：US20240190466A1

公开(公告)日：2024-06-13

申请号：US18065419

申请日：2022-12-13

Applicant: Kodiak Robotics, Inc.

Inventor： Derek J. Phillips ,  Collin C. Otis ,  Andreas Wendel ,  Jackson P. Rusch

IPC: B60W60/00 ,  G01S13/86

CPC classification number: B60W60/0011 ,  G01S13/862 ,  G01S13/865 ,  B60W2420/42 ,  B60W2420/52 ,  B60W2420/54 ,  B60W2420/62

Abstract: This disclosure provides systems and methods for controlling a vehicle. The method comprises receiving data from a set of sensors, wherein the data represents objects or obstacles in an environment of the autonomous vehicle; identifying objects or obstacles from the received data; determining multiple sets of attributes of the objects or obstacles, wherein each set of attributes of the objects or obstacles are determined based on data received by an individual sensor; determining a candidate trajectory for the autonomous vehicle based on the multiple sets of attributes of the objects or obstacles; and controlling the autonomous vehicle according to the candidate trajectory.