Predicted moving trajectory processing method and apparatus, and constraint barrier display method and apparatus
Embodiments of this application provides example methods, media, and apparatuses for predicting moving trajectory of a target object. One example method includes obtaining a candidate moving trajectory, where the candidate moving trajectory is a motion trajectory that is of the target object in future period and that is obtained by using a prediction method, and the target object is in a sensing range of an ego vehicle. Environmental information is obtained within the sensing range. A constraint barrier is generated based on the environmental information, where the constraint barrier is used to indicate an area through which the target object is constrained to pass. The candidate moving trajectory is processed based on the constraint barrier.
1 . A method for predicting moving trajectory of a target object, comprising:
obtaining a candidate moving trajectory, wherein the candidate moving trajectory is a motion trajectory of the target object in future period, the candidate moving trajectory is obtained by a prediction method, the target object is within a sensing range of an ego vehicle, and the ego vehicle is different from the target object;
obtaining environmental information within the sensing range;
generating a constraint barrier based on the environmental information, wherein the constraint barrier is used to indicate an area through which the target object is constrained to pass, the constraint barrier comprises at least one of a hard constraint barrier and a soft constraint barrier, the target object is not able to pass through an area indicated by the hard constraint barrier, and the target object is able to pass through an area indicated by the soft constraint barrier;
processing the candidate moving trajectory based on the constraint barrier; and
performing autonomous driving of the ego vehicle based on the processed candidate moving trajectory of the target object,
wherein the obtaining environmental information within the sensing range comprises:
obtaining a state of a traffic signal light on a target road in the sensing range, wherein the target road is a road on which the ego vehicle is located, and the state of the traffic signal light comprises a passable state and an impassable state; and
wherein the generating a constraint barrier based on the environmental information comprises:
removing the constraint barrier in response to determining that the target object violates a traffic rule indicated by the state of the traffic signal light.
2 . The method according to claim 1 , wherein the obtaining environmental information within the sensing range further comprises:
obtaining a road boundary of a target road in the sensing range and drivable area observation information of the target road in the sensing range, wherein the target road is a road on which the ego vehicle is located; and
the generating a constraint barrier based on the environmental information further comprises:
obtaining an occupied road boundary based on the road boundary and the drivable area observation information, wherein the occupied road boundary indicates a road boundary occupied by an obstacle, and
setting the constraint barrier at the occupied road boundary.
3 . The method according to claim 2 , wherein the obtaining an occupied road boundary based on the road boundary and the drivable area observation information comprises:
projecting the drivable area observation information onto the road boundary;
converting grid information in observation information projected onto the road boundary into a grid occupation probability of a corresponding grid, wherein the grid occupation probability indicates a possibility that the corresponding grid is occupied; and
obtaining the occupied road boundary based on the grid occupation probability by using a sampling manner and a preset rule.
4 . The method according to claim 3 , wherein the obtaining the occupied road boundary based on the grid occupation probability by using a sampling manner and a preset rule comprises:
performing sampling at intervals of a preset distance along the road boundary to obtain a plurality of sampling points;
obtaining a grid occupation probability of a grid in which each of the plurality of sampling points is located; and
obtaining the occupied road boundary according to the grid occupation probability of the grid in which each of the plurality of sampling points is located and the preset rule.
5 . The method according to claim 4 , wherein the obtaining the occupied road boundary according to the grid occupation probability of the grid in which each of the plurality of sampling points is located and the preset rule comprises:
determining whether the grid occupation probability of the grid in which each of the plurality of sampling points is located is greater than a preset threshold; and
when a grid occupation probability of a grid in which a preset quantity of consecutive or nonconsecutive sampling points in the plurality of sampling points are located is greater than the preset threshold, determining that the road boundary is the occupied road boundary.
6 . The method according to claim 1 , wherein the obtaining environmental information within the sensing range further comprises:
obtaining a gap on a target road in the sensing range, wherein the target road is a road on which the ego vehicle is located, and the gap is a gap between a front vehicle and a rear vehicle; and
the generating a constraint barrier based on the environmental information further comprises:
calculating a cross probability of the gap, wherein the cross probability indicates a possibility that a moving object on the target road may cross the gap;
obtaining a target vehicle flow based on the cross probability and the gap, wherein the target vehicle flow comprises an area from a front edge of the front vehicle to a rear edge of the rear vehicle in vehicles before and after the gap; and
setting the constraint barrier at the target vehicle flow.
7 . The method according to claim 6 , wherein the calculating a cross probability of the gap comprises:
evaluating whether the vehicles before and after the gap form a vehicle flow to obtain an evaluation result; and
calculating the cross probability of the gap based on the evaluation result.
8 . The method according to claim 1 , wherein
the generating a constraint barrier based on the environmental information further comprises:
setting the constraint barrier at a stop line of a lane corresponding to a traffic signal light in the impassable state.
9 . The method according to claim 8 , wherein the processing the candidate moving trajectory based on the constraint barrier comprises:
determining whether the candidate moving trajectory intersects the constraint barrier; and
when the candidate moving trajectory intersects the constraint barrier, reducing a prediction result probability of the candidate moving trajectory.
10 . The method according to claim 8 , wherein the processing the candidate moving trajectory based on the constraint barrier comprises:
determining whether the candidate moving trajectory intersects the constraint barrier; and
when the candidate moving trajectory intersects the constraint barrier, truncating the candidate moving trajectory at a position at which the candidate moving trajectory intersects the constraint barrier.
11 . The method according to claim 1 , wherein the hard constraint barrier is related to a road boundary and a drivable area observation information.
12 . The method according to claim 1 , wherein the soft constraint barrier is related to a vehicle flow or by a state of traffic signal light.
13 . The method according to claim 1 , wherein the generating a constraint barrier based on the environmental information further comprises:
reserving the constraint barrier in response to determining that the target object follows the traffic rule indicated by the state of the traffic signal light.
14 . A non-transitory computer-readable storage medium, comprising a program, wherein when the program runs on a computer, the computer is enabled to perform operations comprising:
obtaining a candidate moving trajectory, wherein the candidate moving trajectory is a motion trajectory of a target object in future period, the candidate moving trajectory is obtained by a prediction method, the target object is within a sensing range of an ego vehicle, and the ego vehicle is different from the target object;
obtaining environmental information within the sensing range;
generating a constraint barrier based on the environmental information, wherein the constraint barrier is used to indicate an area through which the target object is constrained to pass, wherein the constraint barrier comprises at least one of a hard constraint barrier and a soft constraint barrier, the target object is not able to pass through an area indicated by the hard constraint barrier, and the target object is able to pass through an area indicated by the soft constraint barrier;
processing the candidate moving trajectory based on the constraint barrier; and
performing autonomous driving of the ego vehicle based on the processed candidate moving trajectory of the target object,
wherein the obtaining environmental information within the sensing range comprises;
obtaining a state of a traffic signal light on a target road in the sensing range, wherein the target road is a road on which the ego vehicle is located, and the state of the traffic signal light comprises a passable state and an impassable state; and
wherein the generating a constraint barrier based on the environmental information comprises:
removing the constraint barrier in response to determining that the target object violates a traffic rule indicated by the state of the traffic signal light.
15 . The non-transitory computer-readable storage medium according to claim 14 , wherein the obtaining environmental information within the sensing range further comprises:
obtaining a road boundary of a target road in the sensing range and drivable area observation information of the target road in the sensing range, wherein the target road is a road on which the ego vehicle is located; and
the generating a constraint barrier based on the environmental information further comprises:
obtaining an occupied road boundary based on the road boundary and the drivable area observation information, wherein the occupied road boundary indicates a road boundary occupied by an obstacle, and
setting the constraint barrier at the occupied road boundary.
16 . The non-transitory computer-readable storage medium according to claim 15 , wherein the obtaining an occupied road boundary based on the road boundary and the drivable area observation information comprises:
projecting the drivable area observation information onto the road boundary;
converting grid information in observation information projected onto the road boundary into a grid occupation probability of a corresponding grid, wherein the grid occupation probability indicates a possibility that the corresponding grid is occupied; and
obtaining the occupied road boundary based on the grid occupation probability by using a sampling manner and a preset rule.
17 . The non-transitory computer-readable storage medium according to claim 16 , wherein the obtaining the occupied road boundary based on the grid occupation probability by using a sampling manner and a preset rule comprises:
performing sampling at intervals of a preset distance along the road boundary to obtain a plurality of sampling points;
obtaining a grid occupation probability of a grid in which each of the plurality of sampling points is located; and
obtaining the occupied road boundary according to the grid occupation probability of the grid in which each of the plurality of sampling points is located and the preset rule.
18 . An apparatus for predicting moving trajectory of a target object, comprising:
one or more processors; and
a non-transitory computer readable medium storing a program to be executed by the one or more processors, wherein the program comprises instructions that when executed by the one or more processors, cause the apparatus to perform operations comprising:
obtaining a candidate moving trajectory, wherein the candidate moving trajectory is a motion trajectory of the target object in future period, the candidate moving trajectory is obtained by a prediction method, the target object is within a sensing range of an ego vehicle, and the ego vehicle is different from the target object;
obtaining environmental information within the sensing range;
generating a constraint barrier based on the environmental information, wherein the constraint barrier is used to indicate an area through which the target object is constrained to pass, wherein the constraint barrier comprises at least one of a hard constraint barrier and a soft constraint barrier, the target object is not able to pass through an area indicated by the hard constraint barrier, and the target object is able to pass through an area indicated by the soft constraint barrier;
processing the candidate moving trajectory based on the constraint barrier; and
performing autonomous driving of the ego vehicle based on the processed candidate moving trajectory of the target object,
wherein the obtaining environmental information within the sensing range comprises:
obtaining a state of a traffic signal light on a target road in the sensing range, wherein the target road is a road on which the ego vehicle is located, and the state of the traffic signal light comprises a passable state and an impassable state; and
wherein the generating a constraint barrier based on the environmental information comprises:
removing the constraint barrier in response to determining that the target object violates a traffic rule indicated by the state of the traffic signal light.
19 . The apparatus according to claim 18 , wherein the obtaining environmental information within the sensing range further comprises:
obtaining a road boundary of a target road in the sensing range and drivable area observation information of the target road in the sensing range, wherein the target road is a road on which the ego vehicle is located; and
the generating a constraint barrier based on the environmental information further comprises:
obtaining an occupied road boundary based on the road boundary and the drivable area observation information, wherein the occupied road boundary indicates a road boundary occupied by an obstacle, and
setting the constraint barrier at the occupied road boundary.
20 . The apparatus according to claim 19 , wherein the obtaining an occupied road boundary based on the road boundary and the drivable area observation information comprises:
projecting the drivable area observation information onto the road boundary;
converting grid information in observation information projected onto the road boundary into a grid occupation probability of a corresponding grid, wherein the grid occupation probability indicates a possibility that the corresponding grid is occupied; and
obtaining the occupied road boundary based on the grid occupation probability by using a sampling manner and a preset rule.