CPC B60W 60/00274 (2020.02) [B60W 30/0956 (2013.01); B60W 30/18159 (2020.02); B60W 40/04 (2013.01); B60W 60/0011 (2020.02); B60W 60/0015 (2020.02); G06N 5/01 (2023.01); G06N 7/01 (2023.01); B60W 2556/50 (2020.02)] | 16 Claims |
1. A vehicle control method, comprising:
acquiring real-time road condition information about an ego vehicle;
acquiring vehicle state information in response to the real-time road condition information indicates that the ego vehicle arrives at a target intersection, wherein the vehicle state information comprises first vehicle state information of the ego vehicle and second vehicle state information of an object vehicle in a predetermined region encompassing the target intersection and a predetermined area of inbound and outbound lanes of the target intersection adjacent the target intersection, the object vehicle comprising an environment vehicle actually passing through the target intersection and two virtual vehicles, the two virtual vehicles being provided at boundaries of a current observation field of view of the ego vehicle, and being configured for simulating potential movement obstructions within blocked areas;
obtaining a prediction result by inputting the vehicle state information into a vehicle motion prediction model; and
controlling the ego vehicle according to the prediction result when the ego vehicle passes through the target intersection.
|