US 12,434,740 B1
Determining object orientation based on parameter modes
Michael Carsten Bosse, Templeton, CA (US); John Bryan Carter, Upton, MA (US); Shuangting Liu, Foster City, CA (US); Francesco Papi, Oakland, CA (US); Nicholas George Dilip Roy, Needham, MA (US); and Zachary Sun, San Francisco, CA (US)
Assigned to Zoox, Inc., Foster City, CA (US)
Filed by Zoox, Inc., Foster City, CA (US)
Filed on Dec. 22, 2022, as Appl. No. 18/087,601.
Int. Cl. B60W 60/00 (2020.01); B60W 40/02 (2006.01)
CPC B60W 60/0027 (2020.02) [B60W 40/02 (2013.01); B60W 2520/14 (2013.01); B60W 2554/4041 (2020.02); B60W 2554/4044 (2020.02); B60W 2554/80 (2020.02)] 20 Claims
OG exemplary drawing
 
1. One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, perform operations comprising:
determining, based at least in part on sensor data generated by at least a sensor of a vehicle, a plurality of parameter values for a current physical parameter associated with an object in an environment;
executing a machine-learned (ML) model to determine, based at least in part on the plurality of parameter values, a parameter value of the plurality of parameter values corresponding to a mode associated with the current physical parameter associated with the object and a probability associated with the parameter value;
wherein the mode associated with the current physical parameter is a common possible parameter value of the current physical parameter;
determining, based at least in part on the parameter value and the probability, a probability distribution associated with the mode;
determining, based at least in part on the probability distribution, a prediction for the object; and
controlling, based at least in part on the prediction, the vehicle.