Method and apparatus for trajectory planning, storage medium, and electronic device

1. A method for trajectory planning, comprising:

determining a preliminary reference trajectory point of a target device and predicted trajectory points of designated obstacles at a target moment;

for each of the designated obstacles, determining a constraint boundary corresponding to the designated obstacle according to a velocity of the designated obstacle and a velocity of the target device, wherein the constraint boundary is configured to divide a space in which the target device and the designated obstacle are located into two disjoint parts, such that the designated obstacle is in one of the two disjoint parts and the target device is in another of the two disjoint parts;

determining a space comprising the target device and enclosed by constraint boundaries corresponding to at least some of the designated obstacles as a constraint set corresponding to the target moment; and

adjusting a preliminary reference trajectory of the target device under the constraint of the constraint set of the target moment, and determining the adjusted preliminary reference trajectory as a reference trajectory of the target device, wherein determining the constraint boundary corresponding to the designated obstacle according to the velocity of the designated obstacle and the velocity of the target device comprises:

determining a relative velocity between the designated obstacle and the target device based on the velocity of the designated obstacle and velocity of the target device;

determining whether a distance between the designated obstacle and the target device has a decreasing trend under an effect of the relative velocity between the designated obstacle and the target device;

in response to determining that the distance between the designated obstacle and the target device has a decreasing trend,

determining hyperplanes each with a normal vector perpendicular to a velocity direction of the designated obstacle as candidate hyperplanes corresponding to the designated obstacle; and

selecting a constraint boundary corresponding to the designated obstacle from the candidate hyperplanes corresponding to the designated obstacle; and

in response to determining that the distance between the designated obstacle and the target device has an increasing trend,

determining hyperplanes each with a normal vector perpendicular to the velocity direction of the target device as candidate hyperplanes corresponding to the designated obstacle; and

selecting a constraint boundary corresponding to the designated obstacle from the candidate hyperplanes corresponding to the designated obstacle controlling vehicle to the updated trajectory.

2. The method according to claim 1 , wherein before determining predicted trajectory points of the designated obstacles, the method further comprises:

determining predicted trajectory points of obstacles; and

selecting obstacles meeting a first distance condition from the obstacles as the designated obstacles based on the preliminary reference trajectory point of the target device, wherein for each of the obstacles meeting the first distance condition, a distance between the predicted trajectory point of the obstacle and the preliminary reference trajectory point of the target device is less than a predetermined first distance.

3. The method according to claim 1 , wherein before for each of the designated obstacles, determining the constraint boundary corresponding to the designated obstacle according to the velocity of the designated obstacle and the velocity of the target device, the method further comprises:

determining, based on the preliminary reference trajectory point of the target device, a space comprising the target device, enclosed by initial constraint boundaries, and centered around the preliminary reference trajectory point of the target device as an initial constraint set; and

determining the space comprising the target device and enclosed by constraint boundaries corresponding to at least some of the designated obstacles as the constraint set corresponding to the target moment comprises:

determining a space comprising the target device and enclosed by at least some of the initial constraint boundaries and the constraint boundaries corresponding to the designated obstacles as a constraint set corresponding to the target moment.

4. The method according to claim 1 , wherein selecting the constraint boundary corresponding to the designated obstacle from the candidate hyperplanes corresponding to the designated obstacle comprises:

for each of the candidate hyperplanes corresponding to the designated obstacle, determining a distance between the candidate hyperplane and the preliminary reference trajectory point of the target device based on the preliminary reference trajectory point of the target device; and

selecting, from the candidate hyperplanes corresponding to the designated obstacle, a candidate hyperplane with a longest distance to the preliminary reference trajectory point of the target device as the constraint boundary corresponding to the designated obstacle.

5. The method according to claim 1 , wherein after determining the predicted trajectory points of the designated obstacles at the target moment, the method further comprises:

determining a taboo list, and initializing the taboo list to an empty set; and

for each of the designated obstacles, determining the constraint boundary corresponding to the designated obstacle according to the velocity of the designated obstacle and the velocity of the target device comprises:

for each of the designated obstacles,

determining the designated obstacle as a target obstacle and determining whether the target obstacle is in the taboo list;

in response to determining that the target obstacle is in the taboo list, reselecting one of the designated obstacles as the target obstacle until all the designated obstacles are comprised in the taboo list; and

in response to determining that the target obstacle is not in the taboo list,

determining the constraint boundary corresponding to the target obstacle based on the velocity of the target obstacle and the velocity of the target device,

adding, based on the constraint boundary corresponding to the target obstacle, one or more of the designated obstacles that are not located in the another of the two disjoint parts, into the taboo list, and

reselecting one of the designated obstacles as the target obstacle.

6. The method according to claim 5 , wherein for each of the designated obstacles, determining the designated obstacle as the target obstacle comprises:

for each of the designated obstacles, determining a distance between the designated obstacle and the target device based on the preliminary reference trajectory point of the target device and the predicted trajectory point of the designated obstacle; and

sorting the designated obstacles according to distances between the designated obstacles and the target device in ascending order, and selecting a target obstacle from the sorted designated obstacles in order.

7. The method according to claim 1 , wherein determining the preliminary reference trajectory point of the target device and the predicted trajectory points of the designated obstacles at the target moment comprises:

for each future moment, determining the future moment as the target moment, and determining the preliminary reference trajectory point of the target device and the predicted trajectory points of the designated obstacles at the target moment; and

adjusting the preliminary reference trajectory of the target device under the constraint of the constraint set of the target moment comprises:

for each future moment, adjusting the preliminary reference trajectory point of the target device at the future moment under the constraint of a constraint set corresponding to the future moment, and determining the adjusted preliminary reference trajectory point as a reference trajectory point of the target device at the future moment; and

determining the reference trajectory of the target device based on reference trajectory points of the target device at future moments.

8. A non-transitory computer-readable storage medium, storing a computer program such that when the computer program is executed by a processor, the processor is caused to perform:

determining a preliminary reference trajectory point of a target device and predicted trajectory points of designated obstacles at a target moment;

for each of the designated obstacles, determining a constraint boundary corresponding to the designated obstacle according to a velocity of the designated obstacle and a velocity of the target device, wherein the constraint boundary is configured to divide a space in which the target device and the designated obstacle are located into two disjoint parts, such that the designated obstacle is in one of the two disjoint parts and the target device is in another of the two disjoint parts;

determining a space comprising the target device and enclosed by constraint boundaries corresponding to at least some of the designated obstacles as a constraint set corresponding to the target moment; and

adjusting a preliminary reference trajectory of the target device under the constraint of the constraint set of the target moment, and determining the adjusted preliminary reference trajectory as a reference trajectory of the target device, wherein determining the constraint boundary corresponding to the designated obstacle according to the velocity of the designated obstacle and the velocity of the target device comprises:

determining a relative velocity between the designated obstacle and the target device based on the velocity of the designated obstacle and velocity of the target device;

determining whether a distance between the designated obstacle and the target device has a decreasing trend under an effect of the relative velocity between the designated obstacle and the target device;

in response to determining that the distance between the designated obstacle and the target device has a decreasing trend,

determining hyperplanes each with a normal vector perpendicular to a velocity direction of the designated obstacle as candidate hyperplanes corresponding to the designated obstacle; and

selecting a constraint boundary corresponding to the designated obstacle from the candidate hyperplanes corresponding to the designated obstacle; and

in response to determining that the distance between the designated obstacle and the target device has an increasing trend,

determining hyperplanes each with a normal vector perpendicular to the velocity direction of the target device as candidate hyperplanes corresponding to the designated obstacle; and

selecting a constraint boundary corresponding to the designated obstacle from the candidate hyperplanes corresponding to the designated obstacle controlling vehicle to the updated trajectory.

9. An unmanned device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor, when executing the program, implementing operations:

determining a preliminary reference trajectory point of a target device and predicted trajectory points of designated obstacles at a target moment;

for each of the designated obstacles, determining a constraint boundary corresponding to the designated obstacle according to a velocity of the designated obstacle and a velocity of the target device, wherein the constraint boundary is configured to divide a space in which the target device and the designated obstacle are located into two disjoint parts, such that the designated obstacle is in one of the two disjoint parts and the target device is in another of the two disjoint parts;

determining a space comprising the target device and enclosed by constraint boundaries corresponding to at least some of the designated obstacles as a constraint set corresponding to the target moment; and

adjusting a preliminary reference trajectory of the target device under the constraint of the constraint set of the target moment, and determining the adjusted preliminary reference trajectory as a reference trajectory of the target device, wherein determining the constraint boundary corresponding to the designated obstacle according to the velocity of the designated obstacle and the velocity of the target device comprises:

determining a relative velocity between the designated obstacle and the target device based on the velocity of the designated obstacle and velocity of the target device;

determining whether a distance between the designated obstacle and the target device has a decreasing trend under an effect of the relative velocity between the designated obstacle and the target device;

in response to determining that the distance between the designated obstacle and the target device has a decreasing trend,

determining hyperplanes each with a normal vector perpendicular to a velocity direction of the designated obstacle as candidate hyperplanes corresponding to the designated obstacle; and

selecting a constraint boundary corresponding to the designated obstacle from the candidate hyperplanes corresponding to the designated obstacle; and

in response to determining that the distance between the designated obstacle and the target device has an increasing trend,

determining hyperplanes each with a normal vector perpendicular to the velocity direction of the target device as candidate hyperplanes corresponding to the designated obstacle; and

selecting a constraint boundary corresponding to the designated obstacle from the candidate hyperplanes corresponding to the designated obstacle controlling vehicle to the updated trajectory.

10. The unmanned device according to claim 9 , wherein before determining predicted trajectory points of the designated obstacles, the operations further comprise:

determining predicted trajectory points of obstacles; and

selecting obstacles meeting a first distance condition from the obstacles as the designated obstacles based on the preliminary reference trajectory point of the target device, wherein for each of the obstacles meeting the first distance condition, a distance between the predicted trajectory point of the obstacle and the preliminary reference trajectory point of the target device is less than a predetermined first distance.

11. The unmanned device according to claim 9 , wherein before for each of the designated obstacles, determining the constraint boundary corresponding to the designated obstacle according to the velocity of the designated obstacle and the velocity of the target device, the operations further comprise:

determining, based on the preliminary reference trajectory point of the target device, a space comprising the target device, enclosed by initial constraint boundaries, and centered around the preliminary reference trajectory point of the target device as an initial constraint set; and

determining the space comprising the target device and enclosed by constraint boundaries corresponding to at least some of the designated obstacles as the constraint set corresponding to the target moment comprises:

determining a space comprising the target device and enclosed by at least some of the initial constraint boundaries and the constraint boundaries corresponding to the designated obstacles as a constraint set corresponding to the target moment.

12. The unmanned device according to claim 9 , wherein selecting the constraint boundary corresponding to the designated obstacle from the candidate hyperplanes corresponding to the designated obstacle comprises:

for each of the candidate hyperplanes corresponding to the designated obstacle, determining a distance between the candidate hyperplane and the preliminary reference trajectory point of the target device based on the preliminary reference trajectory point of the target device; and

selecting, from the candidate hyperplanes corresponding to the designated obstacle, a candidate hyperplane with a longest distance to the preliminary reference trajectory point of the target device as the constraint boundary corresponding to the designated obstacle.

13. The unmanned device according to claim 9 , wherein after determining the predicted trajectory points of the designated obstacles at the target moment, the operations further comprise:

determining a taboo list, and initializing the taboo list to an empty set; and

for each of the designated obstacles, determining the constraint boundary corresponding to the designated obstacle according to the velocity of the designated obstacle and the velocity of the target device comprises:

for each of the designated obstacles,

determining the designated obstacle as a target obstacle and determining whether the target obstacle is in the taboo list;

in response to determining that the target obstacle is in the taboo list, reselecting one of the designated obstacles as the target obstacle until all the designated obstacles are comprised in the taboo list; and

in response to determining that the target obstacle is not in the taboo list,

determining the constraint boundary corresponding to the target obstacle based on the velocity of the target obstacle and the velocity of the target device,

adding, based on the constraint boundary corresponding to the target obstacle, one or more of the designated obstacles that are not located in the another of the two disjoint parts, into the taboo list, and

reselecting one of the designated obstacles as the target obstacle.

14. The unmanned device according to claim 13 , wherein for each of the designated obstacles, determining the designated obstacle as the target obstacle comprises:

for each of the designated obstacles, determining a distance between the designated obstacle and the target device based on the preliminary reference trajectory point of the target device and the predicted trajectory point of the designated obstacle; and

sorting the designated obstacles according to distances between the designated obstacles and the target device in ascending order, and selecting a target obstacle from the sorted designated obstacles in order.

15. The unmanned device according to claim 9 , wherein determining the preliminary reference trajectory point of the target device and the predicted trajectory points of the designated obstacles at the target moment comprises:

for each future moment, determining the future moment as the target moment, and determining the preliminary reference trajectory point of the target device and the predicted trajectory points of the designated obstacles at the target moment; and

adjusting the preliminary reference trajectory of the target device under the constraint of the constraint set of the target moment comprises:

for each future moment, adjusting the preliminary reference trajectory point of the target device at the future moment under the constraint of a constraint set corresponding to the future moment, and determining the adjusted preliminary reference trajectory point as a reference trajectory point of the target device at the future moment; and

determining the reference trajectory of the target device based on reference trajectory points of the target device at future moments.