公开(公告)号：US20230294737A1

公开(公告)日：2023-09-21

申请号：US18187059

申请日：2023-03-21

Applicant: TuSimple, Inc.

Inventor： Zhujia Shi ,  Scott Douglas Foster ,  Dishi Li ,  Hunter Scott Willoughby ,  Charles A. Price ,  Panqu Wang ,  Xiangchen Zhao ,  Yuchao Jin

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

CPC classification number: B60W60/0016 ,  B60W40/02 ,  B60W30/09 ,  B60W2554/80

Abstract: A control subsystem, method, computer program product and autonomous vehicle are provided to respond to an unknown object on a carriageway along which an autonomous vehicle is to travel. In this regard, sensor data is received from at least one vehicle sensor. The sensor data includes location coordinates of the object on the carriageway. The sensor data is evaluated to determine whether the object is known or unknown. If unknown, the size of the object is determined and a motion plan is defined for the autonomous vehicle depending upon the size of the object and the location coordinates of the object relative to the autonomous vehicle. The motion plan that is defined is dependent upon the size of the object with different motion plans being defined for differently sized objects. Driving instructions for the autonomous vehicle are then updated based upon the motion plan that is defined.

公开(公告)号：US10962979B2

公开(公告)日：2021-03-30

申请号：US15721797

申请日：2017-09-30

Applicant: TuSimple, Inc.

Inventor： Xiangchen Zhao ,  Tian Li ,  Panqu Wang ,  Pengfei Chen

IPC: G01C22/00 ,  G05D1/00 ,  G05D1/02 ,  G06K9/00 ,  G06K9/62

Abstract: A system and method for multitask processing for autonomous vehicle computation and control includes: receiving training image data from a training image data collection system; performing a training phase to train a plurality of tasks associated with features of the training image data, the training phase including extracting common features from the training image data, causing the plurality of tasks to generate task-specific predictions based on the training image data, determining a bias between the task-specific prediction for each task and corresponding task-specific ground truth data, and adjusting parameters of each of the plurality of tasks to cause the bias to meet a pre-defined confidence level; receiving image data from an image data collection system associated with an autonomous vehicle; and performing an operational phase including extracting common features from the image data, causing the plurality of trained tasks to concurrently generate task-specific predictions based on the image data.