Uber Technologies, Inc.

Uatc, LLC

Benjamin Kroop

William Payne Ross

Andrew Heine

David McAllister Bradley

Matthew James Way

A system can analyze a live sensor view of a self-driving vehicle (SDV) in accordance with a safety threshold to detect objects of interest along a route and classify each detected object of interest. When the safety threshold is not met, the system can transmit a teleassistance inquiry using LIDAR data to a backend computing system. When a certainty threshold is not met for an object of interest, the system can transmit a different teleassistance inquiry using image data to the backend computing system.

Teleassistance data prioritization for self-driving vehicles

An autonomous vehicle can implement a primary motion planner to continuously determine an first motion plan for the AV, and a secondary motion planner to continuously determine a backup motion plan for the AV. The secondary motion planner can comprise one or more cost metrics that act to diverge the backup motion plan from the first motion plan. A control system of the AV may then analyze a live sensor view generated by a sensor suite of the AV to operate acceleration, braking, and steering systems of the AV along sequential route trajectories selected between the first motion plan and the backup motion plan.

Hierarchical motion planning for autonomous vehicles

A self-driving vehicle can operate by analyzing a live sensor view to autonomously operate acceleration, braking, and steering systems of the SDV along a current route. Based on a detected anomaly associated with the live sensor view, the SDV can transmit a teleassistance inquiry to a backend transport system in accordance with a data prioritization scheme. The SDV can receive a resolution response from the backend transport system based on the teleassistance inquiry, and proceed in accordance with the resolution response.

A self-driving vehicle (SDV) can analyze a live sensor view to autonomously operate acceleration, braking, and steering systems of the SDV along a current route. The SDV can identify an indeterminate object in the live sensor view, and encoding sensor data identifying the indeterminate object for transmission to a backend transport system over one or more networks. The SDV may then receive a resolution response from the backend transport system to resolve the indeterminate object, and cause the SDV to proceed in accordance with the resolution response.

Growing community of inventors

Pittsburgh, PA, United States of America

Benjamin Kroop