Zoox, Inc.

Kshitij Agarwal

David Martins Belo

Andrew Scott Crego

Gerrit Bagschik

Johannes Andreas Huennekens

Thomas Bernard Gacka

Michael Isaac Mizrahi

Julia A Wyatt

Gonzalo Javier Rey

Matthew Nathaniel Dickerson

Pengzhi Tian

Techniques are discussed herein for generating and evaluating autonomous vehicle control systems. In examples, a driving scenario is determined and log data, e.g., generated from user-supported vehicles travelling in an environment, may be used to determine instances in which vehicles experience the driving scenario. Based on the log data, an exposure metric, a severity metric, and/or a controllability metric may be determined. A safety goal can be determined from the metric(s) and vehicle control parameters are determined to comport with the safety goal. The vehicle control parameters may be used to control an autonomous vehicle in the environment.

Safe vehicle operation using log data

Techniques are discussed herein for detecting and measuring faults within vehicle drive systems. In various implementations, actuator feedback monitors and/or vehicle performance monitors are used individually or in combination to determine faults within acceleration systems, braking systems, and/or steering systems of a vehicle. Different monitors may use different algorithms, models, and input data from the vehicle during operation, to compare a predicted drive system output to a corresponding measured output. Additionally, monitors may include associated sets of disable conditions used to distinguish drive system faults from other driving conditions. The outputs from actuator feedback monitors, vehicle performance monitors, and disable conditions may be analyzed to detect drive system faults and control the vehicle to address the faults and control the safe operation of the vehicle.

Model-based vehicle drive system monitors

Techniques for determining a safety margin by which to limit trajectory(ies) generated by a vehicle control system such that the vehicle will not exceed the safety margin more than a target occurrence rate. The techniques may include determining a first spectrum associated with trajectory data generated by one or more vehicles, generating a model of a vehicle, and determining a spectrum of an error signal based at least in part on the model and the first spectrum. Determining the safety margin may be based at least in part on the spectrum of the error signal and a target occurrence rate. Operation characteristics of components of the vehicle (e.g., controller, steering actuator) may be tuned based at least in part on the model, first spectrum, and/or second spectrum. The techniques enable determining safety margins for untested vehicles and/or for different operating states of a vehicle.

Vehicle trajectory controller safety margin

A trajectory for a vehicle can be generated using a lateral offset bias. The vehicle, such as an autonomous vehicle (AV), may be directed to follow reference trajectory for through an environment. The AV may determine a segment associated with the reference trajectory based on curvatures of the reference trajectory, determine a lateral offset bias associated with the segment based at least in part on, for example, one or more of a speed or acceleration of the vehicle, and determine a candidate trajectory for the autonomous vehicle based at least in part on the lateral offset bias. The candidate trajectory may then be used to control the autonomous vehicle.

Trajectory generation using lateral offset biasing

Growing community of inventors

Redwood City, CA, United States of America

Kshitij Agarwal