US 11,734,095 B2
System and method for handling errors in a vehicle neural network processor
Christopher Hsiong, San Jose, CA (US); Emil Talpes, San Mateo, CA (US); Debjit Das Sarma, San Jose, CA (US); Peter Bannon, Woodside, CA (US); Kevin Hurd, Redwood City, CA (US); and Benjamin Floering, San Jose, CA (US)
Assigned to Tesla, Inc., Austin, TX (US)
Filed by Tesla, Inc., Austin, TX (US)
Filed on Sep. 23, 2021, as Appl. No. 17/448,598.
Application 17/448,598 is a continuation of application No. 16/834,104, filed on Mar. 30, 2020, granted, now 11,132,245.
Application 16/834,104 is a continuation of application No. 15/817,005, filed on Nov. 17, 2017, granted, now 10,606,678, issued on Mar. 31, 2020.
Prior Publication US 2022/0083412 A1, Mar. 17, 2022
Int. Cl. G06F 11/00 (2006.01); G06F 11/07 (2006.01); G06N 3/02 (2006.01); G06F 11/32 (2006.01); G06F 11/30 (2006.01)
CPC G06F 11/0772 (2013.01) [G06F 11/0721 (2013.01); G06F 11/0751 (2013.01); G06F 11/3006 (2013.01); G06F 11/3055 (2013.01); G06F 11/327 (2013.01); G06N 3/02 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A neural network processor for executing a plurality of neural networks associated with use of a vehicle, the neural network processor comprising: a plurality of error detectors configured to detect respective errors associated with execution of the neural networks; and a neural network controller configured to receive information from the error detectors and modify operation of the neural network processor based on the received information, wherein the neural network controller maintains one or more status registers for each neural network included in at least a subset of the plurality of neural networks.