Vehicle control method, vehicle control device, and storage medium

1. A vehicle control method, comprising:

recognizing an object present around a vehicle;

generating one or a plurality of target trajectories along which the vehicle is to travel on the basis of the recognized object;

automatically controlling driving of the vehicle on the basis of the generated target trajectory;

calculating a region between a first virtual line, which passes through a reference point using the vehicle as a reference and a first point present in the vicinity of an outer edge of the recognized object, and a second virtual line, which passes through the reference point and a second point present in the vicinity of the outer edge of the object and different from the first point, as a traveling avoidance region that is a region through which traveling of the vehicle should be avoided, and excluding the target trajectory present in the calculated traveling avoidance region from the one or the plurality of target trajectories that were generated; and

automatically controlling the driving of the vehicle on the basis of a remaining target trajectory remained without being excluded.

2. The vehicle control method according to claim 1 , further comprising:

optimizing shapes of the first virtual line and the second virtual line on the basis of model prediction control.

3. The vehicle control method according to claim 1 , further comprising:

calculating a risk region that is a region of a risk distributed around the object, and

inputting the risk region to a model that determines the target trajectory according to the risk region, and generating the one or the plurality of target trajectories on the basis of an output result of the model to which the risk region has been input.

4. The vehicle control method according to claim 3 , wherein the model is a machine learning model learned to output the target trajectory when the risk region is input.

5. The vehicle control method according to claim 3 , further comprising:

setting a position at which a potential of the risk is lower than a threshold in the vicinity of the outer edge of the object in the risk region as the first point and the second point.

6. The vehicle control method according to claim 1 , wherein, when the object recognized is a preceding vehicle that is traveling in front of the vehicle in a lane on which the vehicle is present, a setting of a region behind the preceding vehicle as the traveling avoidance region is not performed.

7. The vehicle control method according to claim 1 , wherein, when a lane change of the vehicle is performed, a calculation of the traveling avoidance region, and an exclusion of the target trajectory present in the traveling avoidance region from the one or the plurality of target trajectories that were generated, are performed.

8. A vehicle control device, comprising:

a hardware processor executing software, hardware including circuitry, or a cooperation of the software and the hardware configured to operate as:

a recognition part configured to recognize an object present around a vehicle;

a generating part configured to generate one or a plurality of target trajectories along which the vehicle is to travel on the basis of the object recognized by the recognition part; and

a driving controller configured to automatically control driving of the vehicle on the basis of the target trajectory generated by the generating part,

wherein the generating part calculates a region between a first virtual line, which passes through a reference point using the vehicle as a reference and a first point present in the vicinity of an outer edge of the object recognized by the recognition part, and a second virtual line, which passes through the reference point and a second point present in the vicinity of the outer edge of the object and different from the first point, as a traveling avoidance region that is a region through which traveling of the vehicle should be avoided, and excludes the target trajectory present in the calculated traveling avoidance region from the one or the plurality of target trajectories that were generated, and

the driving controller automatically controls driving of the vehicle on the basis of a remaining target trajectory remained without being excluded by the generating part.

9. A non-transitory computer-readable storage medium on which a program is stored to execute a computer mounted on a vehicle to:

recognize an object present around the vehicle;

generate one or a plurality of target trajectories along which the vehicle is to travel on the basis of the recognized object;

automatically control driving of the vehicle on the basis of the generated target trajectory;

calculate a region between a first virtual line, which passes through a reference point using the vehicle as a reference and a first point present in the vicinity of an outer edge of the recognized object, and a second virtual line, which passes through the reference point and a second point present in the vicinity of the outer edge of the object and different from the first point, as a traveling avoidance region that is a region through which traveling of the vehicle should be avoided, and exclude the target trajectory present in the calculated traveling avoidance region from the one or the plurality of target trajectories that were generated; and

automatically control the driving of the vehicle on the basis of a remaining target trajectory remained without being excluded.

10. A vehicle control method, comprising:

recognizing an object present around a vehicle;

automatically controlling driving of the vehicle on the basis of at least the recognized object;

calculating a region between a first virtual line, which passes through a reference point using the vehicle as a reference and a first point present in the vicinity of an outer edge of the recognized object, and a second virtual line, which passes through the reference point and a second point present in the vicinity of the outer edge of the object and different from the first point, as a traveling avoidance region that is a region through which traveling of the vehicle should be avoided;

optimizing shapes of the first virtual line and the second virtual line on the basis of model prediction control; and

automatically controlling the driving of the vehicle to avoid traveling through the traveling avoidance region.