US 12,330,692 B2
Route planning system and method of self-driving vehicle
Ji-Fan Yang, Lugang (TW); Tsung-Ming Hsu, Lugang (TW); and Shou-Ti Cheng, Lugang (TW)
Assigned to AUTOMOTIVE RESEARCH & TESTING CENTER, Lugang (TW)
Filed by Automotive Research & Testing Center, Lugang (TW)
Filed on Dec. 27, 2022, as Appl. No. 18/089,089.
Prior Publication US 2024/0208547 A1, Jun. 27, 2024
Int. Cl. B60W 60/00 (2020.01); B60W 50/00 (2006.01); G01S 7/48 (2006.01); G01S 17/89 (2020.01)
CPC B60W 60/0027 (2020.02) [B60W 50/0097 (2013.01); G01S 7/4802 (2013.01); G01S 17/89 (2013.01); B60W 2420/408 (2024.01); B60W 2552/53 (2020.02); B60W 2554/404 (2020.02)] 18 Claims
OG exemplary drawing
 
1. A route planning system of a self-driving vehicle, installed in a host vehicle and comprising:
at least one lidar sensor, used to detect environment information of the host vehicle;
an aerial view generation module, connected with the at least one lidar sensor to receive the environment information and convert the environment information into an aerial view that includes coordinate information of coordinate points;
a feature recognition module, connected with the aerial view generation module, recognizing traffic lane boundaries, traffic lane markings, and other vehicles according to the coordinate information, and marking the traffic lane boundaries, the traffic lane markings, and the other vehicles;
a lane center calculation module, connected with the feature recognition module, working out a driving-allowed region and a lane center according to the traffic lane boundaries and the traffic lane markings marked on the aerial view, finding out a front vehicle from the other vehicles according to the lane center and a position of the host vehicle, and calculating a speed of the front vehicle according to a position of the front vehicle;
a front vehicle prediction module, connected with the feature recognition module and the lane center calculation module, and working out a predicted route of the front vehicle according to a vehicular kinematics model; and
a route planning module, connected with the front vehicle prediction module, and calculating a final route of the host vehicle via using the front vehicle as a route reference point if the predicted route of the front vehicle is the same as a driving route of the host vehicle or via using the traffic lane boundary as a route reference line if the predicted route of the front vehicle is different from the driving route of the host vehicle or if there is no front vehicle,
wherein the coordinate information includes intensity values of return waves of the coordinate points, and
wherein the feature recognition module performs a filtering operation on the coordinate information to filter out noise signals and recognizes the traffic lane boundaries, the traffic lane markings, and the other vehicles in the aerial view according to the intensity values of return waves of the coordinate information.