公开(公告)号：US20240104757A1

公开(公告)日：2024-03-28

申请号：US18339831

申请日：2023-06-22

Applicant: TORC Robotics, Inc.

Inventor： Joseph STAMENKOVICH

IPC: G06T7/60 ,  G01C21/00 ,  G06V10/774 ,  G06V10/82 ,  G06V20/56 ,  G06V20/70

CPC classification number: G06T7/60 ,  G01C21/3815 ,  G01C21/3837 ,  G06V10/774 ,  G06V10/82 ,  G06V20/588 ,  G06V20/70 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30256

Abstract: A method, comprises identifying a set of image data captured by at least one autonomous vehicle when the at least one autonomous vehicle was positioned in a lane of a roadway, and respective ground truth localization data of the at least one autonomous vehicle; determining a plurality of lane width values for the set of image data; labeling the set of image data with the plurality of lane width values, the plurality of lane width values representing a width of a lane in which the at least one autonomous vehicle was positioned; and training using the labeled set of image data, a machine learning model, such that the machine learning model is configured to predict a new lane width value for a new lane as output.

公开(公告)号：US20240101147A1

公开(公告)日：2024-03-28

申请号：US18303460

申请日：2023-04-19

Applicant: TORC Robotics, Inc.

Inventor： Ryan CHILTON ,  Harish PULLAGURLA ,  Joseph STAMENKOVICH

IPC: B60W60/00 ,  G06V10/774 ,  G06V20/56

CPC classification number: B60W60/001 ,  G06V10/774 ,  G06V20/588 ,  B60W2556/40

Abstract: Systems and methods for training and executing machine learning models to generate lane index values are disclosed. A method includes identifying a set of image data captured by at least one autonomous vehicle when the at least autonomous vehicle is positioned in a lane of a roadway and respective ground truth localization data; determining a plurality of lane index values for the set of image data based on the ground truth localization data; labeling the set of image data with the plurality of lane index values, the lane index values representing a number of lanes from a leftmost or rightmost lane to the lane in which the at least one autonomous vehicle was positioned; and training, using the labeled set of image data, a plurality of machine learning models that generate a left lane index value and a right lane index value as output.

公开(公告)号：US20240104938A1

公开(公告)日：2024-03-28

申请号：US18303456

申请日：2023-04-19

Applicant: TORC Robotics, Inc.

Inventor： Ryan CHILTON ,  Harish PULLAGURLA ,  Joseph STAMENKOVICH

IPC: G06V20/56 ,  G01S19/47 ,  G06V10/774

CPC classification number: G06V20/588 ,  G01S19/47 ,  G06V10/774

Abstract: Systems and methods for training and executing machine learning models to generate lane index values are disclosed. A method includes identifying a set of image data captured by at least one autonomous vehicle when the at least autonomous vehicle is positioned in a lane of a roadway and respective ground truth localization data; determining a plurality of lane index values for the set of image data based on the ground truth localization data; labeling the set of image data with the plurality of lane index values, the lane index values representing a number of lanes from a leftmost or rightmost lane to the lane in which the at least one autonomous vehicle was positioned; and training, using the labeled set of image data, a plurality of machine learning models that generate a left lane index value and a right lane index value as output.

公开(公告)号：US20250002044A1

公开(公告)日：2025-01-02

申请号：US18342132

申请日：2023-06-27

Applicant: TORC Robotics, Inc.

Inventor： Joseph STAMENKOVICH

IPC: B60W60/00

Abstract: Systems and methods of automatic correction of map data for autonomous vehicle navigation are disclosed. An autonomous vehicle system can receive first sensor data from a first sensor of an autonomous vehicle and second sensor data from a second sensor of the autonomous vehicle, the first sensor data and the second sensor data captured during operation of the autonomous vehicle; generate, based on the first sensor data, a first prediction of a dimension of a lane of a road; generate, based on the second sensor data, a second prediction of a position of the autonomous vehicle within the lane of the road; determine a confidence value for the second prediction of the position of the autonomous vehicle within the lane of the road based on the first prediction and the second prediction; and navigate the autonomous vehicle based on the confidence value.

公开(公告)号：US20240104939A1

公开(公告)日：2024-03-28

申请号：US18339819

申请日：2023-06-22

Applicant: TORC Robotics, Inc.

Inventor： Joseph STAMENKOVICH

IPC: G06V20/56 ,  G01C21/00 ,  G06V10/774 ,  G06V10/82 ,  G06V20/70

CPC classification number: G06V20/588 ,  G01C21/3822 ,  G01C21/3837 ,  G06V10/774 ,  G06V10/82 ,  G06V20/70

Abstract: A method comprises identifying a set of image data captured by at least one autonomous vehicle when the at least one autonomous vehicle was positioned in a lane of a roadway, and respective ground truth localization data of the at least one autonomous vehicle; determining a total number of lanes for the roadway; labeling the set of image data with the total number of lanes for the roadway; and training using the labeled set of image data, a machine learning model, such that the machine learning model is configured to predict a new total number of lanes for a new roadway as output.

公开(公告)号：US20240101146A1

公开(公告)日：2024-03-28

申请号：US18303449

申请日：2023-04-19

Applicant: TORC Robotics, Inc.

Inventor： Ryan CHILTON ,  Harish PULLAGURLA ,  Joseph STAMENKOVICH

IPC: B60W60/00 ,  G06V10/774 ,  G06V20/56

CPC classification number: B60W60/001 ,  G06V10/774 ,  G06V20/588 ,  B60W2300/14 ,  B60W2552/53 ,  B60W2556/40

Abstract: Systems and methods for training and executing machine learning models to generate lane index values are disclosed. A method includes identifying a set of image data captured by at least one autonomous vehicle when the at least autonomous vehicle is positioned in a lane of a roadway and respective ground truth localization data; determining a plurality of lane index values for the set of image data based on the ground truth localization data; labeling the set of image data with the plurality of lane index values, the lane index values representing a number of lanes from a leftmost or rightmost lane to the lane in which the at least one autonomous vehicle was positioned; and training, using the labeled set of image data, a plurality of machine learning models that generate a left lane index value and a right lane index value as output.