US 12,459,535 B1
Autonomous vehicle anomalous event detection
Christopher Paul Urmson, Los Altos, CA (US); Sterling J. Anderson, Sunnyvale, CA (US); and J. Andrew Bagnell, Pittsburgh, PA (US)
Assigned to Aurora Operations, Inc., Pittsburgh, PA (US)
Filed by Aurora Operations, Inc., Pittsburgh, PA (US)
Filed on Dec. 30, 2024, as Appl. No. 19/005,093.
Application 19/005,093 is a continuation of application No. 19/003,508, filed on Dec. 27, 2024.
Claims priority of provisional application 63/738,993, filed on Dec. 26, 2024.
Int. Cl. B60W 60/00 (2020.01); B60W 50/00 (2006.01)
CPC B60W 60/001 (2020.02) [B60W 50/00 (2013.01); B60W 2050/0028 (2013.01); B60W 2556/45 (2020.02)] 22 Claims
OG exemplary drawing
 
1. A method implemented by one or more processors, the method comprising:
receiving perception data captured by one or more perception sensors of an autonomous vehicle during operation of the autonomous vehicle in an environment;
initiating querying of a generative model using the received perception data to identify an event occurring in the environment during operation of the autonomous vehicle;
in response to identifying the event, automatically initiating a teleassist session between the autonomous vehicle and a remote teleassist system;
periodically initiating querying of the generative model using the communicated perception data; and
dynamically adjusting a rate of the periodic queries based on an operational state of the autonomous vehicle.
 
21. A system comprising:
at least one processor; and
memory storing instructions that, when executed, cause the at least one processor to be operable to perform a method comprising:
receiving perception data captured by one or more perception sensors of an autonomous vehicle during operation of the autonomous vehicle in an environment;
initiating querying of a generative model using the received perception data to identify an event occurring in the environment during operation of the autonomous vehicle, wherein initiating querying of the generative model using the received perception data includes:
initiating querying of a first generative model resident in the autonomous vehicle to generate response data indicating an anomalous event; and
initiating querying of a second generative model remote from the autonomous vehicle and using the response data to attempt to identify the anomalous event; and
in response to identifying the event, automatically initiating a teleassist session between the autonomous vehicle and a remote teleassist system.