US 12,371,010 B2
Integrated trajectory forecasting, error estimation, and vehicle handling when detecting an observed scenario
John Lepird, Pittsburgh, PA (US); Pragati Satpute, Canton, MI (US); and Ramadev Burigsay Hukkeri, Pittsburgh, PA (US)
Assigned to Ford Global Technologies, LLC, Dearborn, MI (US)
Filed by FORD GLOBAL TECHNOLOGIES, LLC, Dearborn, MI (US)
Filed on Dec. 28, 2021, as Appl. No. 17/563,886.
Prior Publication US 2023/0202470 A1, Jun. 29, 2023
Int. Cl. B60W 30/095 (2012.01); B60W 30/09 (2012.01); B60W 40/04 (2006.01); B60W 50/00 (2006.01); B60W 60/00 (2020.01); G06N 3/02 (2006.01); G06N 3/0442 (2023.01); G06N 3/045 (2023.01); G06N 3/09 (2023.01); G06N 5/01 (2023.01); G06N 20/00 (2019.01)
CPC B60W 30/0956 (2013.01) [B60W 30/09 (2013.01); B60W 40/04 (2013.01); B60W 50/0097 (2013.01); B60W 60/0011 (2020.02); B60W 60/00274 (2020.02); G06N 3/02 (2013.01); B60W 2554/4041 (2020.02); B60W 2556/10 (2020.02)] 17 Claims
OG exemplary drawing
 
1. A computer implemented method comprising:
monitoring, by one or more computing devices of an autonomous vehicle (AV), an object within a vicinity of the AV;
generating, by the one or more computing devices, a plurality of trajectories predicting paths the object will take at a future time, the plurality of trajectories being based on a plurality of inputs indicating current and past characteristics of the object;
generating, by the one or more computing devices and with a learned model, a forecasted position of the object at an instance in time based on the plurality of trajectories;
storing, by the one or more computing devices, an error value representing how accurate the forecasted position is versus an observed position of the object, the error value being based on the learned model and one or more heuristic rules;
accumulating, by the one or more computing devices, a plurality of the error values over a period of time;
assigning, by the one or more computing devices, a risk factor for the object based on an average of the plurality of the error values and a predefined correlation associating multiple values of an average error value to a respective value for the risk factor; and
performing, by the one or more computing devices, a maneuver for the AV based on the risk factor.