US 12,233,917 B2
Vehicular autonomous control system based on learned and predicted vehicle motion
Patrick R. Barragan, Cambridge, MA (US); and Charles A. Richter, Cambridge, MA (US)
Assigned to Magna Electronics Inc., Auburn Hills, MI (US)
Filed by Magna Electronics Inc., Auburn Hills, MI (US)
Filed on Jan. 17, 2023, as Appl. No. 18/155,088.
Claims priority of provisional application 63/266,879, filed on Jan. 18, 2022.
Prior Publication US 2023/0227073 A1, Jul. 20, 2023
Int. Cl. B60W 60/00 (2020.01); G06V 10/82 (2022.01); G06V 20/58 (2022.01)
CPC B60W 60/0027 (2020.02) [G06V 10/82 (2022.01); G06V 20/58 (2022.01); B60W 2420/403 (2013.01); B60W 2554/40 (2020.02)] 22 Claims
OG exemplary drawing
 
1. A vehicular vision system, the vehicular vision system comprising:
a forward-viewing camera disposed at a vehicle equipped with the vehicular vision system and viewing at least forward of the equipped vehicle, the camera capturing image data;
wherein the forward-viewing camera comprises a CMOS imaging array, and wherein the CMOS imaging array comprises at least one million photosensors arranged in rows and columns;
an electronic control unit (ECU) comprising electronic circuitry and associated software;
wherein the electronic circuitry of the ECU comprises an image processor for processing image data captured by the forward-viewing camera;
wherein the vehicular vision system, responsive to processing at the ECU of image data captured by the forward-viewing camera, detects a target vehicle forward of the equipped vehicle;
wherein the vehicular vision system, responsive to detecting the target vehicle, predicts, using a machine learning model, a probability for each action of a determined set of actions, and wherein each action in the determined set of actions represents a determined potential action by the target vehicle, and wherein the machine learning model comprises at least one discrete latent variable and at least one continuous latent variable; and
wherein the vehicular vision system at least in part controls the equipped vehicle based at least in part on the predicted probability for each action of the determined set of actions.