Vehicle control system, vehicle control method and vehicle control program

A vehicle control method comprising:

• recognizing, by an in-vehicle computer, a position of a vehicle;
• generating, by the in-vehicle computer, a trajectory which includes a plurality of future target positions to be reached by the vehicle, the plurality of future target positions being consecutively aligned in time series;
• deriving, by the in-vehicle computer, a target speed when the vehicle is caused to travel along the trajectory based on a distance between the plurality of target positions included in the trajectory;
• deriving, by the in-vehicle computer, a first correction amount based on a first deviation between the recognized position of the vehicle and a first target position corresponding to a recognition time at which the recognition of the position of the vehicle has been performed among the plurality of target positions;
• extracting, by the in-vehicle computer, a second target position, which corresponds to a future time after a first predetermined time has elapsed from the recognition time;
• deriving, by the in-vehicle computer, a speed obtained by dividing a length of the trajectory from the first target position to the second target position by the first predetermined time;
• and
• correcting, by the in-vehicle computer, the target speed based on a second deviation between a third target position corresponding to a future time after a second predetermined time has elapsed from the recognition time and a predicted position that the vehicle is predicted to reach at the future time, the second predetermined time being shorter that the first predetermined time.

A non-transitory computer-readable medium comprising a vehicle control program that causes an in-vehicle computer to:

• recognize a position of a vehicle;
• generate a trajectory which includes a plurality of future target positions to be reached by the vehicle, the plurality of future target positions being consecutively aligned in time series;
• derive a target speed when the vehicle is caused to travel along the trajectory based on a distance between the plurality of target positions included in the trajectory;
• derive a first correction amount based on a first deviation between the recognized position of the vehicle and a first target position corresponding to a recognition time at which the recognition of the position of the vehicle has been performed among the plurality of target positions;
• extract a second target position, which corresponds to a future time after a first predetermined time has elapsed from the recognition time, from among the plurality of target positions;
• derive, as the target speed, a speed obtained by dividing a length of the trajectory from the first target position to the second target position by the first predetermined time;
• and
• correct the target speed based on a second deviation between a third target position corresponding to a future time after a second predetermined time has elapsed from the recognition time and a predicted position that the vehicle is predicted to reach at the future time, the second predetermined time being shorter that the first predetermined time.

A vehicle control system, comprising:

• a processor configured to:
• recognize a position of a vehicle;
• generate a trajectory which includes a plurality of future target positions to be reached by the vehicle, the plurality of future target positions being consecutively aligned in time series;
• derive a target speed when the vehicle is caused to travel along the trajectory based on a distance between the plurality of target positions included in the trajectory, and derive a first correction amount based on a first deviation between the position of the vehicle and a first target position corresponding to a recognition time at which the recognition of the position of the vehicle has been performed among the plurality of target positions,
• extract a second target position, which corresponds to a future time after a first predetermined time has elapsed from the recognition time, from among the plurality of target positions,
• derive, as the target speed, a speed obtained by dividing a length of the trajectory from the first target position to the second target position by the first predetermined time,
• derive a third target position corresponding to a future time which is after a second predetermined time has elapsed from the recognition time, the second predetermined time being shorter than the first predetermined time,
• derive a predicted position which is a position in which the vehicle is predicted to reach at the future time which is after the second predetermined time has elapsed from the position of the vehicle when the vehicle was recognized,
• derive a second correction amount based on a second deviation between the third target position and the predicted position, and
• correct the target speed by adding the first correction amount and the second correction amount to the target speed.