Passable space model establishment, road point checking method and device, storage medium and terminal

Abstract

The application discloses a passable space model establishment method, a passable space model point checking method and device, a storage medium and a terminal, wherein the passable space model establishment method comprises the following steps: step S1: generating a voronoi diagram; and step S2: establishing a passable space based on a voronoi point set. On the basis of the above steps, the passable space model point checking method is to judge the distance between a point and each voronoi point in the voronoi point set. The passable space model established based on the voronoi point set can effectively reduce the storage problem caused by the traditional method, and the provided point checking method can avoid the limitation caused by the traditional image slice loading.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, in particular to a passable space model establishment method and device, a storage medium and a terminal. BACKGROUND

[0002] In recent years, with the rapid development of artificial intelligence technology, automatic driving is gradually entering the field of vision of people, and its landing scene is slowly developing in a diversified direction. The core module of the whole automatic driving technology is divided into positioning, perception, decision-making, planning and control, and the core work of the planning module is to plan a collision-free path according to the current environmental information. According to the range of input information, planning can be divided into global path planning and local path planning.

[0003] For global path planning, constructing a passable map space model for path collision detection can ensure the safety and effectiveness of the path to a certain extent, and it is also the core work of the early stage of global path planning. Therefore, how to design a good passable space model for the planning module has very important significance. SUMMARY

[0004] In order to solve the related technical problems, the embodiments of the present application provide a passable space model establishment method and device, a storage medium and a terminal, and the specific technical solutions are as follows:

[0005] In a first aspect, a passable space model establishment method is provided, including the following steps:

[0006] Step S1: generation of a voronoi diagram;

[0007] Step S2: establishment of a passable space based on a voronoi point set.

[0008] In the step S1, the generation of the voronoi diagram is based on the brushfire algorithm.

[0009] The step S1 includes the following steps:

[0010] Step S1.1: Assigning a distance transform value of the non-passable point to a first value;

[0011] Step S1.2: Assigning a distance transform square value of the neighbor grid point of the non-passable point to a second value, and so on, updating the distance transform square value layer by layer outward, wherein the second value is the square of the distance from the non-passable point;

[0012] Step S1.3 voronoi point marking step: in the updating process, the crossed and overlapped grid points of different impassable point diffusion appear, which are marked as voronoi points, and the distance transform square values are stored.

[0013] The step S2 comprises the following steps:

[0014] Step S2.1 step of obtaining voronoi point set;

[0015] Step S2.2 model establishing step: taking each voronoi point as the center and the corresponding distance transform value as the radius to make a circle.

[0016] The second aspect provides a passable space model road point checking method, the passable space model is established by the passable space model establishing method as above,

[0017] The passable space model road point checking method is to judge the distance between the road point and each voronoi point in the voronoi point set.

[0018] The step S2 comprises the following steps:

[0019] S11 step of finding the nearest voronoi point: taking the road point to be checked as the reference, finding the nearest voronoi point and recording its distance transform value as the third value;

[0020] S12 step of judging whether the road point is in the passable space: calculating the distance between the road point and the nearest voronoi point as the fourth value, if the fourth value is less than the third value, the road point is in the passable space; if the fourth value is greater than the third value, the road point is not in the passable space.

[0021] The third aspect provides a passable space model establishing device, comprising the following units:

[0022] The voronoi diagram generating unit,

[0023] The voronoi diagram generating unit is used for the following actions:

[0024] The distance transform value of the impassable point is set as the first value;

[0025] The distance transform square value of the neighborhood grid point of the impassable point is set as the second value, and the distance transform square value is updated layer by layer outward, wherein the second value is the square of the distance from the impassable point;

[0026] In the updating process, the crossed and overlapped grid points diffused by different impassable points are marked as voronoi points, and the distance transform square values are stored;

[0027] The establishing unit for the passable space based on the voronoi point set,

[0028] The establishing unit for the passable space based on the voronoi point set is used for the following actions:

[0029] The step for obtaining the voronoi point set;

[0030] The step for taking each voronoi point as a center and the corresponding distance transform value as a radius to draw a circle.

[0031] In a fourth aspect, a passable space model road point checking device is provided, and the passable space model includes the passable space model establishing device and the checking unit,

[0032] The checking unit is used for checking the passable space model road point by judging the distance between the road point and each voronoi point in the voronoi point set.

[0033] In a fifth aspect, a storage medium is provided, and the storage medium stores at least one instruction, at least one program, a code set or an instruction set. The at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to realize the method as described above.

[0034] In a sixth aspect, a terminal is provided, and the terminal includes a processor and a memory. The memory stores at least one instruction, at least one program, a code set or an instruction set. The at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to realize the method as described above.

[0035] Compared with the prior art, the passable space model based on the voronoi point set provided by the present application can effectively reduce the storage problem caused by the traditional method. In addition, the road point checking method provided by the present application can avoid the limitations caused by the traditional image slice loading. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 The passable space model establishing method flowchart of the embodiment of the present application;

[0037] Figure 2 The passable space model road point checking method flowchart of the embodiment of the present application;

[0038] Figure 3 The brushfire algorithm schematic diagram of the embodiment of the present application;

[0039] Figure 4 A schematic diagram of the voronoi diagram generation process for the embodiment of the present application;

[0040] Figure 5 A schematic diagram of the establishment of the passable space model for the embodiment of the present application;

[0041] Figure 6 A schematic diagram of the case where the waypoint is in the passable space for the embodiment of the present application;

[0042] Figure 7 A schematic diagram of the case where the waypoint is outside the passable space for the embodiment of the present application. DETAILED DESCRIPTION

[0043] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0044] The present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0045] Conceptual description:

[0046] Voronoi diagram: also known as Thiessen polygon or Dirichlet diagram, which is composed of a series of continuous polygons composed of vertical bisectors connecting two adjacent points. N points that are different in the plane, according to the nearest neighbor principle to divide the plane; each point is associated with its nearest neighbor region.

[0047] As shown in Figure 1 , the passable space model establishment method provided by the embodiment of the present application comprises the following steps:

[0048] Step S1: Generation of voronoi diagram;

[0049] It should be noted that the passable space model proposed by the present application is realized based on the voronoi diagram, and the passable space is constructed according to the voronoi point.

[0050] The generation of the voronoi diagram can use the method commonly used in the industry, or the method in the present application. The present application proposes a relatively novel brushfire algorithm to construct the voronoi diagram.

[0051] As shown in Figure 3 , Figure 4 , the specific steps are as follows:

[0052] Step S1.1: Assigning the distance transform value of the impassable point to the first value;

[0053] It should be noted that, for the whole map, the distance transform value of the non-passable point is set to a first value from the non-passable area, wherein the first value can be set to 0;

[0054] Step S1.2: Assign values to grid points: set the distance transform square value of the neighborhood grid points of the non-passable point to a second value, and so on, update the distance transform square value of the grid points layer by layer outward, wherein the second value is the square of the distance from the non-passable point;

[0055] It should be noted that, after assigning values to the non-passable point, the 8-neighborhood grid points around the non-passable point are then traversed in turn, and the distance transform square value (which stores the square of the distance from the center obstacle) of the traversed grid points is updated in turn to a second value, which is associated with the first value, and so on, the grid points are updated layer by layer outward in a wave-like structure.

[0056] Step S1.3: voronoi point marking step: during the updating process, the grid points spread out by different non-passable points intersect and overlap, indicating that the grid points at this place can exist as a dividing point between obstacle points, and are marked as voronoi points, and their distance transform square values are also stored.

[0057] Step S2: Establishing a passable space based on the voronoi point set.

[0058] As shown in Figure 5 , it specifically includes the following steps:

[0059] Step S2.1: Obtain the voronoi point set;

[0060] It should be noted that, according to step S1, after generating the voronoi diagram, a series of voronoi points can be obtained, and the distance transform square value is also obtained, so the distance change value of each voronoi point, i.e. the distance from the nearest obstacle point, can be obtained, and thus a series of voronoi point sets {Voro(x, y, dis)} are obtained. Storing voronoi points will obviously save a lot of space compared to storing the whole map, so the passable space model can be established through the voronoi point set next.

[0061] Step S2.2: Model establishment step: take each voronoi point as the center and the corresponding distance transform value as the radius to make a circle.

[0062] It should be noted that, since the distance transform value in each voronoi point represents the distance to the nearest impassable point, and the generated voronoi diagram itself can represent the skeleton structure of the entire map, the space constructed by taking each voronoi point as the center and the corresponding distance transform value as the radius can roughly represent the entire passable space.

[0063] In addition, the embodiment of the present application also provides a waypoint verification method based on the above passable space model, wherein the passable space model is established by the above passable space model establishment method.

[0064] The passable space model waypoint verification method is to judge the distance between the waypoint and each voronoi point in the voronoi point set.

[0065] As shown in Figure 2 , the passable space model waypoint verification method specifically includes the following steps:

[0066] S11, a step of finding the nearest voronoi point: taking the waypoint to be verified as the reference, finding the nearest voronoi point and recording its distance transform value as a third value;

[0067] S12, a step of judging whether the waypoint is in the passable space: calculating the distance between the waypoint and the nearest voronoi point as a fourth value, if the fourth value is less than the third value, the waypoint is in the passable space; if the fourth value is greater than the third value, the waypoint is not in the passable space.

[0068] It should be noted that, in the passable space model constructed by the voronoi diagram, the waypoint verification can be realized by judging the distance between the waypoint and each voronoi point.

[0069] Specifically, taking the waypoint to be verified as the reference (x, y), find the nearest voronoi point (x_voro, y_voro) from the voronoi point set to the waypoint to be verified, and the corresponding distance transform value is dis_voro. The way to find the nearest point can be realized by the form of KD_Tree; then calculate the distance between the waypoint and the voronoi point, if the distance is less than dis_voro, the waypoint is in the passable space, if the distance is greater than dis_voro, it is considered that the waypoint is not in the passable space, as shown in Figure 6 and Figure 7 .

[0070] In addition, in the preferred embodiment, corresponding to the above passable space model establishment method, a passable space model establishment device is provided, which includes the following units:

[0071] A voronoi diagram generation unit,

[0072] The voronoi diagram generation unit is used for the following actions:

[0073] The distance transform value of the impassable point is set as a first numerical value;

[0074] The distance transform square value of the neighborhood grid point of the impassable point is set as a second numerical value, and so on, and the distance transform square value is updated layer by layer outwardly, wherein the second numerical value is the square of the distance from the impassable point;

[0075] In the updating process, the grid points diffused by different impassable points intersect and overlap, and are marked as voronoi points, and the distance transform square values are stored;

[0076] A passable space establishment unit based on the voronoi point set,

[0077] The passable space establishment unit based on the voronoi point set is used for the following actions:

[0078] A step for obtaining the voronoi point set;

[0079] A step for taking each voronoi point as the center and the corresponding distance transform value as the radius to draw a circle.

[0080] In addition, in the preferred embodiment, a passable space model road point checking device is provided corresponding to the passable space model road point checking method described above, and the passable space model includes the passable space model establishment device and the checking unit described above,

[0081] The checking unit is used for checking the passable space model road point by judging the distance between the road point and each voronoi point in the voronoi point set.

[0082] In addition, in the preferred embodiment, the present application provides a storage medium, wherein at least one instruction, at least one program, a code set or an instruction set is stored in the storage medium, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to realize the passable space model establishment method and / or the passable space model road point checking method in the above-mentioned embodiments.

[0083] It should be noted that those skilled in the art can understand that all or part of the steps of the above-mentioned embodiments can be completed by hardware, or by a program to instruct related hardware, and the program can be stored in a computer readable storage medium, and the storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0084] In addition, in a preferred embodiment, the present application provides a terminal, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, which is loaded and executed by the processor to implement the passable space model establishing method and / or the pass point checking method of the passable space model in the above-mentioned embodiments.

[0085] It should be noted that the technical solutions not described in detail in the present application adopt known technologies.

[0086] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should also be considered as the protection scope of the present application.

Claims

1. A method of establishing a navigable space model, characterized by, comprising the following steps: Step S1: voronoi diagram generation; The step S1 comprises the following steps: Step S1.1: Assigning distance transform value of the impassable point to a first value; Step S1.2: Assigning distance transform square value of the neighbor grid points of the impassable point to a second value, and so on, updating the distance transform square value of the grid points layer by layer outward, wherein the second value is the square of the distance from the impassable point; Step S1.3: Voronoi point marking step: in the updating process, the grid points diffused by different impassable points intersect and overlap, which are marked as voronoi points, and their distance transform square values are stored; Step S2: Building a passable space based on the voronoi point set.

2. The passable space model building method according to claim 1, wherein In the step S1, the voronoi diagram is generated based on the brushfire algorithm.

3. The passable space model building method according to claim 1, wherein The step S2 comprises the following steps: Step S2.1: Obtaining the voronoi point set; Step S2.2: Model building step: taking each voronoi point as the center and the corresponding distance transform value as the radius to draw a circle.

4. A method for checking a model waypoint of a navigable space, characterized in that The passable space model The passable space model is built by the passable space model building method according to any one of claims 1-3, The passable space model point checking method is to judge the distance between the point and each voronoi point in the voronoi point set.

5. The passable space model point checking method according to claim 4, comprising the following steps: S11: Finding the nearest voronoi point step: taking the point to be checked as the reference, finding the nearest voronoi point and recording its distance transform value as a third value; S12: Judging whether the point is in the passable space step: calculating the distance between the point and the nearest voronoi point as a fourth value, if the fourth value is less than the third value, the point is in the passable space; if the fourth value is greater than the third value, the point is not in the passable space. comprising the following units:

6. A passable space model establishing apparatus characterized by comprising: A voronoi diagram generation unit, The voronoi diagram generation unit is used for the following actions: Assigning distance transform value of the impassable point to a first value; Assigning distance transform square value of the neighbor grid points of the impassable point to a second value, and so on, updating the distance transform square value of the grid points layer by layer outward, wherein the second value is the square of the distance from the impassable point; In the updating process, the grid points diffused by different impassable points intersect and overlap, which are marked as voronoi points, and their distance transform square values are stored; A passable space building unit based on the voronoi point set, The passable space building unit based on the voronoi point set is used for the following actions: A step for obtaining the voronoi point set; ​ For each voronoi point, a circle is drawn with the voronoi point as the center and the corresponding distance transform value as the radius.

7. A navigable space model waypoint verification device, characterized by, The passable space model The passable space model establishment device and the checking unit as claimed in claim 6, The checking unit is configured to check the road points of the passable space model by judging the distance between the road points and each voronoi point in the voronoi point set.

8. A storage medium, characterized by The storage medium stores at least one instruction, at least one program, a code set or an instruction set, which are loaded and executed by the processor to implement the method according to any one of claims 1 to 5.

9. A terminal, characterized by comprising: The terminal comprises a processor and a memory, and the memory stores at least one instruction, at least one program, a code set or an instruction set, which are loaded and executed by the processor to implement the method according to any one of claims 1 to 5.

Images