ROAD CONSTRAINT DETERMINING METHOD AND APPARATUS
This application discloses a road constraint determining method and apparatus, applied to the intelligent driving field, and in particular, to a sensor in an advanced driver assistance system ADAS or an autonomous driving system, for example, radar and/or a photographing apparatus. In this method, a moving state of a target is determined based on detection information of the target; at least one road geometry of a road on which the target is located is determined based on the detection information of the target; and a road constraint of the target is determined based on the at least one road geometry and the moving state of the target. The road constraint includes at least one of a road direction constraint and a road width constraint. According to the solutions in this application, road constraint determining accuracy can be improved, and target tracking accuracy can be further improved.
1 . A road constraint determining method, comprising:
determining a moving state of a target based on detection information of the target;
determining, based on the detection information of the target, at least one road geometry of a road on which the target is located, wherein each of the at least one road geometry is represented by using at least one piece of information; and
determining a road constraint of the target based on the at least one road geometry and the moving state of the target, wherein the road constraint comprises at least one of a road direction constraint or a road width constraint.
2 . The method according to claim 1 , wherein the method further comprises:
determining at least one target road geometry in the at least one road geometry; and
the determining a road constraint of the target based on the at least one road geometry and the moving state of the target comprises:
determining the road constraint of the target based on the at least one target road geometry and the moving state of the target.
3 . The method according to claim 2 , wherein the determining at least one target road geometry in the at least one road geometry comprises:
performing the following steps for each of the at least one road geometry:
determining a tangent direction angle of the road geometry at a first location, wherein the tangent direction angle is an included angle between a tangent line of the road geometry at the first location and a radial direction;
obtaining a tangent direction angle at a target location of the target based on a lateral velocity and a radial velocity at the target location of the target, wherein a distance between the target location of the target and the first location falls within a first distance range; and
determining the road geometry as the target road geometry if an absolute value of a difference between the tangent direction angle at the first location and the tangent direction angle at the target location is less than a first threshold.
4 . The method according to claim 2 , wherein the determining at least one target road geometry in the at least one road geometry comprises:
performing the following steps for each of the at least one road geometry:
determining the road geometry as the target road geometry if a distance between the target and the road geometry falls within a second distance range.
5 . The method according to claim 2 , wherein the determining at least one target road geometry in the at least one road geometry comprises:
performing the following steps for each of the at least one road geometry:
obtaining a distance between the target and the road geometry; and
determining, based on a quantity of the at least one road geometry, that Num road geometries at a smallest distance is the at least one target road geometry, wherein Num is a positive integer not less than 1.
6 . The method according to claim 2 , wherein the determining the road constraint of the target based on the at least one target road geometry and the moving state of the target comprises:
determining at least one second location respectively located in the at least one target road geometry, wherein the at least one second location is a location closest to the target in at least one first target road geometry, and the first target road geometry is a target road geometry in which the second location is located; and
determining the road direction constraint of the target based on a confidence level of the at least one target road geometry and a tangent direction angle of the at least one target road geometry at the at least one second location.
7 . The method according to claim 6 , wherein
the confidence level of the target road geometry is a confidence level of a road parameter of the target road geometry; or
the confidence level of the target road geometry is a confidence level of the tangent direction angle of the target road geometry at the second location.
8 . The method according to claim 7 , wherein when the confidence level of the target road geometry is the confidence level of the road parameter of the target road geometry, the method further comprises:
determining the confidence level of the road parameter of the target road geometry based on a variance or a standard deviation of the road parameter of the target road geometry, wherein the road parameter is at least one piece of information used to represent the target road geometry; or
when the confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second location, the method further comprises:
determining the confidence level of the tangent direction angle of the target road geometry at the second location based on a variance or a standard deviation of the tangent direction angle of the target road geometry at the second location.
9 . The method according to claim 6 , wherein the determining the road direction constraint of the target based on a confidence level of the at least one target road geometry and a tangent direction angle of the at least one target road geometry at the at least one second location comprises:
determining, based on the confidence level of the target road geometry, a weight value that is of the tangent direction angle of the at least one target road geometry at the at least one second location and that exists during fusion; and
determining, based on the weight value, that a fusion result obtained after fusion is performed on the tangent direction angle is the road direction constraint of the target.
10 . The method according to claim 6 , wherein the determining the road direction constraint of the target based on a confidence level of the at least one target road geometry and a tangent direction angle of the at least one target road geometry at the at least one second location comprises:
determining that the tangent direction angle at the second location is the road direction constraint of the target, wherein the second location is a location closest to the target in a second target road geometry, and the second target road geometry is a target road geometry with a highest confidence level in the at least one target road geometry.
11 . A road constraint determining apparatus, comprising: at least one processor and a memory, wherein
the memory is configured to store program instructions; and
the at least one processor is configured to invoke and execute the program instructions stored in the memory, to enable the apparatus to perform the method of:
determining a moving state of a target based on detection information of the target; and
determining, based on the detection information of the target, at least one road geometry of a road on which the target is located, wherein each of the at least one road geometry is represented by using at least one piece of information; and
determining a road constraint of the target based on the at least one road geometry and the moving state of the target, wherein the road constraint comprises at least one of a road direction constraint and a road width constraint.
12 . The apparatus according to claim 11 , wherein the method further comprising:
determining at least one target road geometry in the at least one road geometry; and
determining the road constraint of the target based on the at least one target road geometry and the moving state of the target.
13 . The apparatus according to claim 12 , wherein the method further comprising:
determining a tangent direction angle of the road geometry at a first location, wherein the tangent direction angle is an included angle between a tangent line of the road geometry at the first location and a radial direction;
obtaining a tangent direction angle at a target location of the target based on a lateral velocity and a radial velocity at the target location of the target, wherein a distance between the target location of the target and the first location falls within a first distance range; and
determining the road geometry as the target road geometry if an absolute value of a difference between the tangent direction angle at the first location and the tangent direction angle at the target location is less than a first threshold.
14 . The apparatus according to claim 12 , wherein the method further comprising:
determining the road geometry as the target road geometry if a distance between the target and the road geometry falls within a second distance range.
15 . The apparatus according to claim 12 , wherein the method further comprising:
obtaining a distance between the target and the road geometry; and
determining, based on a quantity of the at least one road geometry, that Num road geometries at a smallest distance is the at least one target road geometry, wherein Num is a positive integer not less than 1.
16 . The apparatus according to claim 12 , wherein the method further comprising:
determining at least one second location respectively located in the at least one target road geometry, wherein the at least one second location is a location closest to the target in at least one first target road geometry, and the first target road geometry is a target road geometry in which the second location is located; and
determining the road direction constraint of the target based on a confidence level of the at least one target road geometry and a tangent direction angle of the at least one target road geometry at the at least one second location.
17 . The apparatus according to claim 16 , wherein
the confidence level of the target road geometry is a confidence level of a road parameter of the target road geometry; or
the confidence level of the target road geometry is a confidence level of the tangent direction angle of the target road geometry at the second location.
18 . The apparatus according to claim 16 , wherein the method further comprising:
determining, based on the confidence level of the target road geometry, a weight value that is of the tangent direction angle of the at least one target road geometry at the at least one second location and that exists during fusion; and
determining, based on the weight value, that a fusion result obtained after fusion is performed on the tangent direction angle is the road direction constraint of the target.
19 . The apparatus according to claim 16 , wherein the method further comprising:
determining that the tangent direction angle at the second location is the road direction constraint of the target, the second location is a location closest to the target in a second target road geometry, and the second target road geometry is a target road geometry with a highest confidence level in the at least one target road geometry.
20 . A computer-readable storage medium, wherein
the computer-readable storage medium stores instructions; and when the instructions are run on a computer, the computer is enabled to perform the method of:
determining a moving state of a target based on detection information of the target;
determining, based on the detection information of the target, at least one road geometry of a road on which the target is located, wherein each of the at least one road geometry is represented by using at least one piece of information; and
determining a road constraint of the target based on the at least one road geometry and the moving state of the target, wherein the road constraint comprises at least one of a road direction constraint or a road width constraint.