US 11,741,719 B2
Approach to maneuver planning for navigating around parked vehicles for autonomous driving
Brent N. Bacchus, Sterling Heights, MI (US); Pinaki Gupta, Novi, MI (US); and Sayyed Rouhollah Jafari Tafti, Troy, MI (US)
Assigned to GM GLOBAL TECHNOLOGY OPERATIONS LLC, Detroit, MI (US)
Filed by GM GLOBAL TECHNOLOGY OPERATIONS LLC, Detroit, MI (US)
Filed on Aug. 27, 2019, as Appl. No. 16/552,686.
Prior Publication US 2021/0061293 A1, Mar. 4, 2021
Int. Cl. G06V 20/58 (2022.01); G05D 1/00 (2006.01); B60W 50/00 (2006.01); G06V 10/764 (2022.01)
CPC G06V 20/58 (2022.01) [B60W 50/0097 (2013.01); G05D 1/0088 (2013.01); G06V 10/764 (2022.01)] 17 Claims
OG exemplary drawing
 
1. A computer-implemented method of navigating an autonomous vehicle, comprising:
classifying, at a processor, each of a plurality of remote stationary vehicles into an object hypothesis, each object hypothesis including a possible location for an associated remote stationary vehicle and a probability associated with the possible location that the associated remote stationary vehicle is in a parked state;
determining, at the processor, a proposed behavior for the autonomous vehicle for each object hypothesis, each proposed behavior based on the possible location for the associated remote stationary vehicle and the probability associated with the possible location;
determining a temporal cost for each proposed behavior based on a longitudinal acceleration of the proposed behavior;
calculating a set cost for each object hypothesis based on a product of the probability associated with the object hypothesis and the temporal cost associated with the object hypothesis;
selecting an actionable behavior for the autonomous vehicle from the proposed behavior for which the set cost is a least value; and
controlling the autonomous vehicle, at the processor, to perform the actionable behavior.