Vehicle control device
In the vehicle control device mounted on the vehicle, the information acquisition unit 11 acquires a vehicle information data group representing vehicle information regarding the behavior of the vehicle 2 , a road environment data group 32 representing road environment information regarding the environment of the road on which the vehicle 2 is traveling, and a sensor recognition data group 33 representing sensor recognition information regarding the surrounding environment of the vehicle 2 recognized by the external sensor group 4 mounted on the vehicle 2 . Based on the road environment data group 32 and the sensor recognition data group 33 , the apparent obstacle prediction unit 12 predicts the future behavior of the apparent obstacle that may hinder the traveling of the vehicle 2 . The lane risk map generation unit 15 generates a lane risk map representing the collision risk between the vehicle 2 and the obstacle for each lane of the road based on the vehicle information data group 31 and the future behavior of the apparent obstacle predicted by the apparent obstacle prediction unit 12.
1 . A vehicle control device mounted on a host vehicle, the vehicle control device comprising:
one or more sensors mounted on the host vehicle and configured to recognize a surrounding environment of the host vehicle; and
a processor configured to:
acquire vehicle information regarding a behavior of the host vehicle, road environment information regarding an environment of a road on which the host vehicle is traveling, and sensor recognition information regarding the surrounding environment of the host vehicle;
predict a future behavior of an apparent obstacle that is determined to obstruct traveling of the host vehicle on a basis of the road environment information and the sensor recognition information;
generate a lane grid map having a plurality of grid points set at equal intervals on a plurality of straight lines representing each lane of the road on a basis of the vehicle information and the road environment information;
generate a lane risk map representing a collision risk between the host vehicle and an obstacle for each lane of the road on a basis of the vehicle information and the predicted future behavior of the apparent obstacle, by calculating the collision risk at each grid point of the lane grid map; and
control an actuator to cause the host vehicle to follow a trajectory based on the lane risk map,
wherein the lane grid map is generated based on normalization processing of changing intervals between positions respectively corresponding to the plurality of grid points on each lane of the road for each lane according to a difference in a path length between lanes due to a difference in a shape of each lane of the road.
2 . The vehicle control device according to claim 1 , wherein the normalization processing includes:
setting a plurality of representative points on a traveling lane of the host vehicle on the road at equal intervals;
setting a plurality of straight lines that are orthogonal to an extension direction of the traveling lane at the plurality of representative points respectively and pass through the plurality of representative points, respectively;
setting intersections between the plurality of straight lines and each lane of the road excluding the traveling lane as a plurality of reference points respectively corresponding to the plurality of representative points; and
recording position information of the representative point and the reference point in association with each grid point of the lane grid map.
3 . The vehicle control device according to claim 1 , wherein the processor is further configured to correct a distance between grid points of the lane grid map on a basis of the difference in the path length, and calculate the collision risk on a basis of the corrected distance.
4 . The vehicle control device according to claim 1 , wherein the processor is further configured to: identify a blind spot region that is not detectable by the one or more sensors around the host vehicle,
predict a potential obstacle that is determined to appear from the blind spot region, and generate the lane risk map on a basis of the vehicle information, future behavior of the apparent obstacle, and a prediction result of the potential obstacle.
5 . The vehicle control device according to claim 4 , wherein the processor is further configured to: generate a lane grid map having a plurality of grid points set at equal intervals on a plurality of straight lines representing each lane of the road on a basis of the vehicle information and the road environment information,
calculate a first collision risk between the host vehicle and the apparent obstacle at each grid point on the lane grid map and a second collision risk between the host vehicle and the potential obstacle at each grid point on the lane grid map, and generate the lane risk map on a basis of calculation results of the first and second collision risks.
6 . The vehicle control device according to claim 5 , wherein the processor is further configured to:
calculate arrival times of the host vehicle, the apparent obstacle, and the potential obstacle at each grid point of the lane grid map,
calculate the first collision risk by comparing an arrival time of the host vehicle with an arrival time of the apparent obstacle, and
calculate the second collision risk by comparing an arrival time of the host vehicle with an arrival time of the potential obstacle.