Granted Patent B2

US 12,304,531 · App. 18/148,525 · Granted May 20, 2025

Inventors: Shinichi Shiraishi (Saitama, JP); Hidetaka Fuse (Tokyo-to, JP)

Assignee: TOYOTA JIDOSHA KABUSHIKI KAISHA

B60W60/005B60W50/0097B60W60/001B60W2420/403B60W2420/408

Abstract

Claims (36)

1. A device for generating a trajectory, comprising a processor configured to:

detect, from each piece of first data, the position of an object in an area around a vehicle at generation of the piece of first data to output first object information indicating the position, wherein the first data is time-series data generated by a first sensor mounted on the vehicle and represents the surroundings of the vehicle,

detect, from each piece of the first data, a predicted position of an object in the area around the vehicle at a predetermined period after generation of the piece of the first data to output first predictive object information indicating the predicted position,

detect, from each piece of second data, the position of an object in an area around the vehicle to output second object information indicating the position, wherein the second data is time-series data generated by a second sensor mounted on the vehicle and differing from the first sensor and represents the surroundings of the vehicle,

predict whether an autonomous driving level at which travel of the vehicle is controlled will change from a first autonomous driving level at which a driver of the vehicle is under first watching obligation to a second autonomous driving level at which the driver is under second watching obligation different from the first watching obligation,

identify, when a change in autonomous driving level is predicted, a changing time of the predicted change, and

generate a trajectory along which the vehicle will travel, using the first object information until the predetermined period before the changing time, using the first predictive object information from the predetermined period before the changing time until the changing time, and using the second object information after the changing time;

wherein

at outputting the first predictive object information, the processor outputs the first object information and the first predictive object information, using time-series environmental images generated by the first sensor depending on light focused by an optical system having a first focal length as the first data; and

at outputting the second object information, the processor outputs the second object information, using time-series environmental images generated by the second sensor depending on light focused by an optical system having a second focal length shorter than the first focal length as the second data.

2. The device according to claim 1 , wherein

at outputting the first predictive object information, the processor is further configured to calculate confidence of the first predictive object information and outputs confidence score information indicating the confidence; and

at generating the trajectory, the processor generates the trajectory, using the first predictive object information and the confidence score information from the predetermined period before the changing time until the changing time.

3. The device according to claim 2 , wherein at generating the trajectory, the processor generates the trajectory, using first predictive object information corresponding to the confidence score information when the confidence indicated by the confidence score information exceeds a confidence score threshold, and without using first predictive object information corresponding to the confidence score information when the confidence indicated by the confidence score information does not exceed the confidence score threshold.

4. The device according to claim 1 , wherein the processor finishes outputting the first predictive object information and starts outputting the second object information, at the predetermined period before the changing time.

5. The device according to claim 4 , wherein

at outputting the first predictive object information, the processor outputs the first object information and the first predictive object information, using the first data when generation of the first data is the predetermined period or more before the changing time; and the processor does not output the first object information or the first predictive object information, using the first data when generation of the first data is not the predetermined period or more before the changing time; and

at outputting the second object information, the processor does not output the second object information, using the second data when generation of the second data is the predetermined period or more before the changing time; and the processor outputs the second object information, using the second data when generation of the second data is not the predetermined period or more before the changing time.

6. A method for generating a trajectory, comprising:

detecting, from each piece of first data, the position of an object in an area around a vehicle at generation of the piece of first data to output first object information indicating the position, wherein the first data is time-series data generated by a first sensor mounted on the vehicle and represents the surroundings of the vehicle;

detecting, from each piece of the first data, a predicted position of an object in the area around the vehicle at a predetermined period after generation of the piece of first data to output first predictive object information indicating the predicted position;

detecting, from each piece of second data, the position of an object in an area around the vehicle to output second object information indicating the position, wherein the second data is time-series data generated by a second sensor mounted on the vehicle and differing from the first sensor and represents the surroundings of the vehicle;

predicting whether an autonomous driving level at which travel of the vehicle is controlled will change from a first autonomous driving level at which a driver of the vehicle is under first watching obligation to a second autonomous driving level at which the driver is under second watching obligation different from the first watching obligation;

identifying, when a change in autonomous driving level is predicted, a changing time of the predicted change;

generating a trajectory along which the vehicle will travel, using the first object information until the predetermined period before the changing time, using the first predictive object information from the predetermined period before the changing time until the changing time, and using the second object information after the changing time;

outputting the first predictive object information, the processor outputs the first object information and the first predictive object information, using time-series environmental images generated by the first sensor depending on light focused by an optical system having a first focal length as the first data; and

outputting the second object information, the processor outputs the second object information, using time-series environmental images generated by the second sensor depending on light focused by an optical system having a second focal length shorter than the first focal length as the second data.

7. A non-transitory computer-readable medium storing a computer program for generating a trajectory, the computer program causing a computer mounted on a vehicle to execute a process comprising:

detecting, from each piece of first data, the position of an object in an area around the vehicle at generation of the piece of first data to output first object information indicating the position, wherein the first data is time-series data generated by a first sensor mounted on the vehicle and represents the surroundings of the vehicle;

detecting, from each piece of the first data, a predicted position of an object in the area around the vehicle at a predetermined period after generation of the piece of first data to output first predictive object information indicating the predicted position;

detecting, from each piece of second data, the position of an object in an area around the vehicle to output second object information indicating the position, wherein the second data is time-series data generated by a second sensor mounted on the vehicle and differing from the first sensor and represents the surroundings of the vehicle;

predicting whether an autonomous driving level at which travel of the vehicle is controlled will change from a first autonomous driving level at which a driver of the vehicle is under first watching obligation to a second autonomous driving level at which the driver is under second watching obligation different from the first watching obligation;

identifying, when a change in autonomous driving level is predicted, a changing time of the predicted change;

generating a trajectory along which the vehicle will travel, using the first object information until the predetermined period before the changing time, using the first predictive object information from the predetermined period before the changing time until the changing time, and using the second object information after the changing time;

outputting the first predictive object information, the processor outputs the first object information and the first predictive object information, using time-series environmental images generated by the first sensor depending on light focused by an optical system having a first focal length as the first data; and

outputting the second object information, the processor outputs the second object information, using time-series environmental images generated by the second sensor depending on light focused by an optical system having a second focal length shorter than the first focal length as the second data.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (11)