An all-coverage path planning method, device and medium for outdoor scenes

Abstract

The application discloses an outdoor scene-oriented full-coverage path planning method and device, and a medium, wherein the method comprises the following steps: determining a region to be covered, dividing the determined region into a plurality of sub-regions according to a preset shape; acquiring the width of the sub-regions, and calculating a dynamic adjustment factor of each sub-region according to the width; merging the sub-regions with the same dynamic adjustment factor and connected to each other, and constructing a coordinate system corresponding to the sub-regions; in each sub-region, constructing a spanning tree with the minimum number of turning times of a coverage path as an optimization target, and obtaining a path with the minimum number of turning times; converting the paths in all the sub-regions into a global map coordinate system and connecting the paths, and obtaining a complete global coverage path. The application can effectively improve the cleaning coverage rate, cleaning efficiency and cleaning quality of a cleaning robot on a cleaning region. The application can be widely applied to the technical field of robots.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robots, and in particular to a full-coverage path planning method for outdoor scenes, a device and a medium. BACKGROUND

[0002] In recent years, more and more service robots are endowed with the task of performing simple service functions, such as cleaning robots, meal delivery robots, etc. How to improve the performance of robots, reduce maintenance requirements and expand the functional range has become an important topic. Cleaning work is a labor-intensive work, and the development of robot technology has promoted the development of cleaning robots. At present, cleaning robots can complete cleaning tasks on structured roads, but there is no mature solution for outdoor semi-structured scenes (sidewalks, parks, etc.). In the outdoor semi-structured scene, the cleaning robot needs to be able to achieve high coverage, low repeatability, and high efficiency in completing the cleaning task. Before the unmanned cleaning robot performs the cleaning task, it needs to pre-construct a map of the cleaning environment and plan a full-coverage cleaning path, so that the cleaning robot can track the path and thus achieve full-coverage cleaning.

[0003] The existing coverage path planning methods are mainly classified as follows:

[0004] (1) Coverage method based on unit decomposition. This method is a classic planning algorithm, which is more suitable for simple or regular static environments. The quality of unit decomposition is directly affected by the complexity of the environment. In a complex environment containing a large number of obstacles, the decomposition calculation may become complex and inefficient, and a large amount of post-processing is required to ensure complete coverage. After completing the unit decomposition, an arch-shaped coverage method needs to be used in each sub-unit to complete the coverage path planning task in the sub-unit, and individual optimization path operations may need to be performed for each unit, increasing the difficulty of algorithm implementation and debugging. After completing the path planning in the unit, the paths of all units need to be connected, and the paths used for connection may cause path repetition, which may increase the cost of repeated shuttle paths between different units.

[0005] (2) Coverage method based on neural network. The coverage method based on neural network has good learning ability and adaptability in complex scenes, especially in irregular environments and dynamic scenes. However, the performance of neural network is highly dependent on the quality and quantity of training data, and outdoor scenes usually have diversity and complexity. If the training data cannot cover these diversities, the neural network may not be able to generalize. For large-scale high-quality coverage tasks, the cost of data collection and labeling is high. Moreover, the training and inference of neural network models require high computing resources, and it is difficult to deploy complex neural network models on performance-limited cleaning robots.

[0006] (3) The spanning tree-based coverage algorithm. The traditional spanning tree-based coverage method often uses a single size grid to discretize the coverable area, which may cause the spanning tree construction to be unable to completely cover the area, resulting in coverage blind spots. And the traditional spanning tree-based coverage method performs well in constructing a simple coverage path, but usually only guarantees coverage and does not optimize path length or task efficiency, which may result in low task efficiency. SUMMARY

[0007] To at least partially solve one of the technical problems existing in the prior art, the purpose of the present application is to provide a full coverage path planning method, device and medium for outdoor semi-structured scenes.

[0008] The first technical solution adopted by the present application is:

[0009] A full coverage path planning method for outdoor scenes, comprising the following steps:

[0010] Determine the area to be covered, and divide the determined area into a plurality of sub-areas according to a predetermined shape;

[0011] Obtain the width of the sub-area, and calculate the dynamic adjustment factor of each sub-area according to the width;

[0012] Merge the sub-areas with the same dynamic adjustment factor and connected to each other, and construct a coordinate system corresponding to the sub-areas;

[0013] In each sub-area, a spanning tree is constructed with the optimization objective of minimizing the number of turns of the coverage path, and a path containing the minimum number of turns is obtained;

[0014] Convert the paths in all sub-areas to the global map coordinate system and connect them to obtain a complete global coverage path.

[0015] Further, the adjustable width of the cleaning robot's coverage tool is taken as the dynamic adjustment factor δ;

[0016] The calculation formula of the dynamic adjustment factor δ is:

[0017]

[0018] In the formula, W is the width of the sub-area, and N is a positive integer.

[0019] Further, the spanning tree is constructed in each sub-area with the optimization objective of minimizing the number of turns of the coverage path, and a path containing the minimum number of turns is obtained, comprising:

[0020] The path planning is performed based on a spanning tree covering method; the path planned based on the spanning tree covering method is a covering path generated from a starting point along branches of the spanning tree and finally returning to the starting point;

[0021] When the spanning tree is constructed in the sub-region, a mathematical model is established with the minimum number of turning times of the covering path as an optimization objective, and a minimum spanning tree is solved; the turning of the covering path only occurs at the end points of the branches or the connection between the branches, and the optimization objective of the mathematical model is converted into solving the minimum number of spanning tree branches and the optimal branch connection mode.

[0022] Further, the conversion of the optimization objective of the mathematical model into solving the minimum number of spanning tree branches and the optimal branch connection mode comprises:

[0023] According to the dynamic adjustment factor, each sub-region is discretized into a grid, and one large grid is composed of four small grids;

[0024] A spanning tree node is arranged at the center of each large grid, and a variable v i is used to represent the spanning tree node, wherein v i ∈{v1,v2,……,v n}, n is the number of large grids; a direction operator DIR(v i )∈{H,V} is introduced for each spanning tree node to represent the extension direction of the branch where the node is located, H represents that the node is in the horizontal direction, and V represents that the node is in the vertical direction;

[0025] Adjacent nodes satisfying a constraint condition are merged one by one to obtain spanning tree branches, and the constraint condition is that: (1) the DIR(v i ) of a node and its horizontally adjacent node are both H; (2) the DIR(v i ) of a node and its vertically adjacent node are both V; therefore, the total number of branches in the coverable region depends on the directions assigned to the nodes; since each spanning tree branch has two end points, in order to simplify the model, solving the minimum number of spanning tree branches is equivalent to solving the minimum number of end points, and only the left end points of the horizontal branches or the upper end points of the vertical branches are calculated.

[0026] Further, the conversion of the optimization objective of the mathematical model into solving the minimum number of spanning tree branches and the optimal branch connection mode comprises:

[0027] L(v i ) is defined as the left node of v i , and T(v i ) is defined as the upper node of v i ; in order to identify the nodes close to the boundary, a boundary variable d i is introduced to represent the boundary, and d iWhen DIR(d i )=V;when d i When indicating the left boundary, DIR(d i )=H;

[0028] Define the endpoint discrimination operator END(v i ) is used to identify node v i Whether it is an endpoint, the expression is as follows:

[0029]

[0030] Where END(v i ) is a binary operator used to determine v i Whether it is the endpoint of a spanning tree branch;

[0031] According to the operator END(v i )A mathematical model is established to solve the problem with the minimum number of endpoints, that is, the minimum number of branches.

[0032] Furthermore, according to the operator END(v i The expression of the mathematical model established is as follows:

[0033]

[0034] Furthermore, the paths in all sub-areas are converted to the global map coordinate system and connected to obtain a complete global coverage path, including:

[0035] After finding the minimum number of spanning tree branches, connect all the branches to get a complete spanning tree: the path generated along the spanning tree branches has two turns at both ends, so the turn cost of all unconnected branches is:

[0036] C branches =4kC t

[0037] In the formula, k represents the minimum number of branches solved, C t Indicates the cost of a single turn is 1;

[0038] The total turn cost of the spanning tree is expressed as:

[0039]

[0040] Where, Represents node v i and v j The change in steering cost after connection, Indicates that the node v i and v j connected edges;

[0041] All branch connections are selected by a greedy strategy to select branch connection mode to generate a tree, so as to minimize the total steering cost;

[0042] All paths are projected into the global coordinate system through a rotation matrix R and a translation matrix T, and are connected to obtain a full coverage path with the least number of steering.

[0043] Further, the steering cost corresponding to any node of the generated tree is represented as:

[0044]

[0045] In the formula, deg(v i ) represents the degree of node v i , that is, the number of adjacent nodes;

[0046] There are five ways to connect any two generated tree branches: L-shaped connection, n-shaped connection, T-shaped connection, h-shaped connection and H-shaped connection; connecting any two branches will change the degrees of connection nodes v i and v j , thereby affecting the total steering cost C of the generated tree; the expression of the change amount is:

[0047]

[0048] The second technical solution adopted by the present application is:

[0049] An electronic device, comprising a processor and a memory, the memory storing 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 being loaded and executed by the processor to implement an outdoor scene-oriented full coverage path planning method as described above.

[0050] The third technical solution adopted by the present application is:

[0051] A computer-readable storage medium, the storage medium storing 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 being loaded and executed by the processor to implement an outdoor scene-oriented full coverage path planning method as described above.

[0052] The fourth technical solution adopted by the present application is:

[0053] A computer program product or computer program comprises computer instructions stored in a computer readable storage medium. A processor of a computer device can read the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to make the computer device perform the method described above.

[0054] The present application has the beneficial effects that: the present application divides a plurality of sub-regions with different geometric characteristics by introducing a dynamic adjustment factor, and constructs a coordinate system corresponding to the sub-regions, so as to improve the overall coverage. In each sub-region, a spanning tree is constructed with the minimum number of turns of the coverage path as the optimization objective, so as to obtain a path with the minimum number of turns. Finally, the paths in all sub-regions are converted to the global map coordinate system and connected to obtain a complete global coverage path. The present application can effectively improve the cleaning coverage, cleaning efficiency and cleaning quality of the cleaning robot on the cleaning area. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following introduces the drawings of the related technical solutions in the embodiments of the present application or the prior art. It should be understood that the drawings in the following introduction are only for the convenience of clearly describing some embodiments in the technical solutions of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0056] Figure 1 is a schematic diagram of sub-region division and merging in the embodiments of the present application; wherein, Figure 1 in (a) is the initially divided sub-region, Figure 1 in (b) is the merged sub-region;

[0057] Figure 2 is a schematic diagram of solving the minimum number of spanning tree branches in the embodiments of the present application;

[0058] Figure 3 is a schematic diagram of the connection mode of the spanning tree branches in the embodiments of the present application; Figure 3 in (a) is L-shaped connection, Figure 3 in (b) is n-shaped connection, Figure 3 in (c) is T-shaped connection, Figure 3 in (d) is h-shaped connection, Figure 3 in (e) is H-shaped connection;

[0059] Figure 4 is a flow chart of the global coverage cleaning path planning method in the embodiments of the present application;

[0060] Figure 5is a step flow chart of a full coverage path planning method for an outdoor scene in an embodiment of the present application. DETAILED DESCRIPTION

[0061] Embodiments of the present application are described in detail below with reference to the accompanying drawings, in which the same or similar elements or elements having the same or similar functions are denoted by the same or similar reference numerals throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application and cannot be understood as limiting the present application. For the step numbers in the following embodiments, they are only set for the convenience of explanation and do not limit the order between the steps, and the execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0062] In the description of the present application, it should be understood that the orientation description, such as the orientation or position relationship indicated by up, down, front, back, left, right, etc. is based on the orientation or position relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0063] In the description of the present application, one or more is meant by several, more than two is meant by more than two, greater than, less than, more than, etc. are not included in the number, above, below, etc. are included in the number. If it is described as first, second, it is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features or the order of indicated technical features.

[0064] In the description of the present application, unless otherwise explicitly limited, the words such as setting, installing, connecting, etc. should be broadly understood, and those skilled in the art can reasonably determine the specific meaning of the above words in the present application in combination with the specific content of the technical solution.

[0065] Embodiment 1

[0066] As shown in Figure 4 and Figure 5 The present embodiment provides a full coverage path planning method for an outdoor scene, specifically comprising the following steps:

[0067] S1, determining the area to be covered, dividing the determined area into a plurality of sub-areas according to a preset shape;

[0068] S2, obtaining the width of the sub-area, calculating the dynamic adjustment factor of each sub-area according to the width;

[0069] S3, merging the sub-regions with the same dynamic adjustment factor and connecting them, and constructing a coordinate system corresponding to the sub-regions;

[0070] S4, in each sub-region, constructing a spanning tree with the minimum number of turning times of the coverage path as the optimization target, and obtaining a path with the minimum number of turning times;

[0071] S5, converting the paths in all sub-regions to the global map coordinate system and connecting them to obtain a complete global coverage path.

[0072] The working principle of the embodiment is: since the cleaning robot has the feature of dynamically adjusting the coverage tool, a dynamic adjustment factor can be introduced according to the dynamic adjustment range of the coverage tool to improve the area coverage rate. The method first divides the region into several sub-regions with different geometric characteristics according to the dynamic adjustment factor. Then, in each sub-region, a spanning tree is constructed with the minimum number of turning times of the path as the target, so that a full coverage path with the minimum number of turning times is obtained in the sub-region. Finally, the coverage path in each sub-region is projected to the global map coordinate system, and all the sub-region paths are connected to obtain a full coverage cleaning path with high coverage rate, few turning times, and no path repetition.

[0073] The above method will be explained in detail in combination with the drawings and specific embodiments.

[0074] Referring to Figure 4 In the cleaning task of the outdoor semi-structured scene, it is required to cover the cleaning area as completely as possible and to reduce the time for completing the cleaning task as much as possible, therefore, the embodiment proposes a full coverage path planning method for the outdoor semi-structured scene. The method mainly includes two parts of calculating the dynamic adjustment factor and sub-region division, and constructing the minimum spanning tree.

[0075] (1) Calculation of dynamic adjustment factor and sub-region division

[0076] Since the outdoor semi-structured region may contain many channels with different widths and inclinations, the embodiment introduces a dynamic adjustment factor δ and divides the sub-regions. First, according to the orientation position of each channel, each channel is divided into a sub-region, as shown in (a) of Figure 1 The dynamic adjustment factor δ of each sub-region is calculated as follows:

[0077]

[0078] 1≤δ≤1.5 (2)

[0079] Where W is the geometric width of the feasible channel, δ can be used to represent the size of the discretized grid, and can also represent the adjustable width of the cleaning robot's covering tool. Merging sub-regions with the same δ can reduce the number of sub-regions and construct a coordinate system for the merged sub-regions, such as Figure 1 As shown in (b), it lays a good foundation for the subsequent coverage path planning task. Figure 1 Before the sub-regions are merged, they include sub-region 1, sub-region 2, sub-region 3, sub-region 4, sub-region 5, and sub-region 6. Since the dynamic adjustment factors δ corresponding to sub-region 1, sub-region 4, sub-region 5, and sub-region 6 are the same, these sub-regions are merged to finally obtain sub-region 1, sub-region 2, and sub-region 3; coordinate systems are constructed for these three sub-regions respectively.

[0080] in Figure 1 The black area in the middle represents an obstacle. In some embodiments, the robot collects environmental information (such as radar information or image information) through sensors, identifies the boundaries of objects based on the collected information, and identifies the area and obstacles that the robot needs to cover. This can be achieved using existing technical means and is not the focus of this patent, so it is not described in detail here. In other embodiments, the environment can be directly input externally without the need for sensor collection.

[0081] (2) Constructing a minimum spanning tree

[0082] Since the path planned based on the spanning tree coverage (STC) method is characterized by a covered path generated from the starting point along the branches of the spanning tree and ultimately returning to the starting point, this path characteristic is consistent with the cleaning task behavior of the cleaning robot. Therefore, the coverage path planning method of this embodiment uses STC as its core and proposes a full coverage cleaning path planning method with high coverage and few turns.

[0083] When constructing a spanning tree in a subregion, the optimization objective is to minimize the number of turns along the covering path. A mathematical model is then developed to solve for the minimum spanning tree. Turns along the covering path occur only at the endpoints of branches or at the junctions between branches. Therefore, the optimization objective of the mathematical model can be transformed into finding the minimum number of spanning tree branches and the optimal branch connection method.

[0084] like Figure 2 As shown, first use the grid to discretize the coverable area. A large grid is composed of four small grids. Assume that the coverable area is discretized into n large grids, and a spanning tree node is set at the center of each large grid. Use the variable v i To represent the spanning tree node, where v i ∈{v1,v2,……,v n}. Introduce a direction operator DIR(vi ) e {H, V} is used to represent the extension direction of the branch where the node is located, H means the node is in horizontal direction, and V means the node is in vertical direction.

[0085] The adjacent nodes satisfying the constraint condition are merged one by one to obtain the spanning tree branch, the constraint condition is: (1) the DIR(v i ) of a node and its horizontal adjacent node are both H; (2) the DIR(v i ) of a node and its vertical adjacent node are both V. Therefore, the total number of branches in the coverable area depends on the directions assigned to the nodes.

[0086] Since each spanning tree branch has two endpoints, in order to simplify the model, solving the minimum number of spanning tree branches is equivalent to solving the minimum number of endpoints. We only calculate the left endpoints of horizontal branches or the upper endpoints of vertical branches. We define L(v i ) as the left node of v i , and T(v i ) as the upper node of v i . In order to identify the nodes close to the boundary, the boundary variable d i is introduced. When d i represents the upper boundary, DIR(d i ) = V; when d i represents the left boundary, DIR(d i ) = H. Further, we can define the endpoint judgment operator END(v i ) to judge whether the node v i is an endpoint, as follows:

[0087]

[0088] where END(v i ) is a binary operator to determine whether v i is an endpoint of a spanning tree branch. Then we use the operator to establish a mathematical model to solve the minimum number of endpoints, that is, the minimum number of branches:

[0089]

[0090]

[0091] After solving the minimum number of spanning tree branches, all branches are connected to obtain the complete spanning tree. The path generated along the spanning tree branch has two turning points at both ends, so the turning cost of all unconnected branches is:

[0092] C branches = 4kC t (6)

[0093] where k denotes the number of the least branches solved, C t denotes the cost of single turning 1.

[0094] The turning cost corresponding to any node of the spanning tree is denoted as:

[0095]

[0096] where deg(v i ) denotes the degree of node v i , i.e., the number of adjacent nodes.

[0097] There are five main ways to connect any two spanning tree branches, i.e., L-shaped connection, n-shaped connection, T-shaped connection, h-shaped connection and H-shaped connection, as shown in Figure 3 Connecting any two branches will change the degrees of the connected nodes v i and v j , thereby affecting the turning cost C of the spanning tree. We define the cost function to represent the amount of change in the turning cost after connecting nodes v i and v j :

[0098]

[0099] where e denotes the edge connecting nodes v i and v j .

[0100] The total turning cost of the spanning tree can be represented as:

[0101]

[0102] The greedy strategy is adopted to select the branch connection mode to connect all branches to obtain a spanning tree, so as to minimize the total turning cost.

[0103] In each sub-region, the above planning method is used to solve a covering path with the least number of turns. Then these paths are projected into the global coordinate system through the rotation matrix R and the translation matrix T. For any path point P i (x, y) in the sub-region coordinate system, its position in the global coordinate system can be obtained by solving the equation Y = RX + T. Convert the path of each sub-region to the global map coordinate system, and finally connect the paths of all sub-regions to obtain a global covering path with the least number of turns.

[0104] Compared with the traditional full coverage path planning method, the full coverage path planning method can effectively improve the coverage rate of the coverage path and significantly reduce the number of turns of the global coverage cleaning path, and is helpful to improve the cleaning quality and efficiency.

[0105] Embodiment 2

[0106] The embodiment of the present application also provides an electronic device, which 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 realize the full coverage path planning method for outdoor scenes as shown in the method. Figure 5

[0107] It can be understood that the memory can include a random access memory (RAM) and a read-only memory (ROM). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory can be used to store instructions, programs, codes, code sets or instruction sets. The memory can include a program storage area and a data storage area, wherein the program storage area can store instructions for implementing an operating system, instructions for at least one function, instructions for implementing various method embodiments described above, etc.; and the data storage area can store data created according to the use of the server, etc.

[0108] The processor can include one or more processing cores. The processor connects various parts in the whole server through various interfaces and lines, executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory, and calling data stored in the memory. Optionally, the processor can be implemented in at least one of the following hardware forms: digital signal processing (DSP), field programmable gate array (FPGA), programmable logic array (PLA). The processor can be integrated with one or several combinations of central processing unit (CPU) and modem. Among them, the CPU mainly processes operating systems and application programs, etc.; and the modem is used to process wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor, but be realized by a single chip.

[0109] ​Since the electronic device is an electronic device corresponding to the full coverage path planning method for outdoor scenarios according to the embodiments of the present application, and the principle of solving problems of the electronic device is similar to that of the method, the implementation of the electronic device can be referred to the implementation process of the above-mentioned method embodiments, and the repeated parts will not be described herein.

[0110] Embodiment 3

[0111] The embodiments of the present application also provide a computer readable 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 implement the full coverage path planning method for outdoor scenarios as shown in Figure 5

[0112] Those skilled in the art can understand that all or part of the steps of the various methods in the above embodiments can be completed by a program instructing relevant hardware, and the program can be stored in a computer readable storage medium, including a Read-Only Memory (ROM), a Random Access Memory (RAM), a Programmable Read-only Memory (PROM), an Erasable Programmable Read Only Memory (EPROM), a One-time Programmable Read-Only Memory (OTPROM), an Electrically-Erasable Programmable Read-Only Memory (EEPROM), a Compact Disc Read-Only Memory (CD-ROM) or other optical disk storage, a magnetic disk storage, a magnetic tape storage, or any other medium that can be used to carry or store data which can be read by a computer.

[0113] Since the storage medium is a storage medium corresponding to the full coverage path planning method for outdoor scenarios according to the embodiments of the present application, and the principle of solving problems of the storage medium is similar to that of the method, the implementation of the storage medium can be referred to the implementation process of the above-mentioned method embodiments, and the repeated parts will not be described herein.

[0114] Embodiment 4

[0115] ​In some possible implementation manners, each aspect of the method of the embodiment of the present application can also be implemented in the form of a program product, which includes program codes for causing a computer device to execute the steps of the method of the present application when the program product is run on the computer device. The executable computer program codes or "codes" for executing each embodiment can be written in a high-level programming language such as C, C++, C#, Smalltalk, Java, JavaScript, Visual Basic, Structured Query Language (e.g., Transact-SQL), Perl, or in various other programming languages.

[0116] It should be understood that various parts of the present application can be realized in hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be realized in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if realized in hardware, and as in another embodiment, any one or a combination of the following technologies known in the art can be used: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0117] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0118] The above embodiments are only for the purpose of illustrating the technical concept and characteristics of the present application, and the purpose is to enable those skilled in the art to understand the content of the present application and to implement it, and cannot limit the protection scope of the present application. Any equivalent changes or modifications made according to the essence of the present application should be covered within the protection scope of the present application.

Claims

1. A full coverage path planning method for outdoor scenarios, characterized in that, The method comprises the following steps: determining a region to be covered, dividing the determined region into a plurality of sub-regions according to a preset shape; acquiring the width of each sub-region and calculating a dynamic adjustment factor of each sub-region according to the width; merging sub-regions with the same dynamic adjustment factor and connecting them, and constructing a coordinate system corresponding to the sub-regions; in each sub-region, constructing a spanning tree with the optimization objective of minimizing the number of turns of the coverage path to obtain a path with the least number of turns; converting the paths in all sub-regions to a global map coordinate system and connecting them to obtain a complete global coverage path; the step of constructing a spanning tree in each sub-region with the optimization objective of minimizing the number of turns of the coverage path to obtain a path with the least number of turns comprises: adopting a spanning tree coverage method for path planning; the path planned by the spanning tree coverage method is a path generated from the starting point along the branches of the spanning tree and finally returning to the starting point; when constructing the spanning tree in the sub-region, a mathematical model is established with the optimization objective of minimizing the number of turns of the coverage path to solve the minimum spanning tree; the turns of the coverage path only occur at the endpoints of the branches or the connections between the branches, and the optimization objective of the mathematical model is converted to solving the minimum number of spanning tree branches and the optimal branch connection mode; the step of converting the optimization objective of the mathematical model to solving the minimum number of spanning tree branches and the optimal branch connection mode comprises: grid discretization is performed on each sub-region according to the dynamic adjustment factor, and one large grid is composed of four small grids; A spanning tree node is set at the center of each macro-grid, denoted by variable , where , is the number of macro-grids; a direction operator is introduced for each spanning tree node to indicate the extending direction of the branch where the node is located, indicates that the node is in the horizontal direction, indicates that the node is in the vertical direction; The adjacent nodes satisfying the constraint condition are merged one by one to obtain the spanning tree branch, and the constraint condition is that: (1) the are all ; (2) the are all ; therefore, the total number of branches in the coverable area depends on the direction assigned to the nodes; since each spanning tree branch has two endpoints, in order to simplify the model, solving the minimum number of spanning tree branches is equivalent to solving the minimum number of endpoints, and only the left endpoints of the horizontal branches or the upper endpoints of the vertical branches are calculated; the step of converting the paths in all sub-regions to a global map coordinate system and connecting them to obtain a complete global coverage path comprises: after solving the minimum number of spanning tree branches, all the branches are connected to obtain a complete spanning tree; the turn cost of all unconnected branches is: In the formula, represents the number of the least branches solved out, represents the cost 1 of single turning; the total turn cost of the spanning tree is represented as: In the formula, representing a node and a change in steering cost after connection, representing a node and an edge connecting the nodes; a greedy strategy is adopted to select the branch connection mode to connect all the branches to obtain the spanning tree, so that the total turn cost is minimized; All paths are projected into the global coordinate system by rotation and translation matrices and connected to obtain a globally shortest path with full coverage. and connected to obtain a globally shortest path with full coverage.

2. The full coverage path planning method for outdoor scenes according to claim 1, wherein, Adjustable width of a robot's covering tool as a dynamic adjustment factor ; The dynamic adjustment factor The calculation formula is: wherein is the width of the sub-region, is a positive integer.

3. The full coverage path planning method for outdoor scenes according to claim 1, wherein, the step of solving the minimum number of spanning tree branches is equivalent to solving the minimum number of endpoints comprises: Definition is the left node of is the upper node of To identify nodes close to the border, introduce the border variable denotes the border, when denotes the upper border, when ; denotes the left border, when ; ; Definition of endpoint predicate operator For determining whether a node is an endpoint, the expression is as follows: whether an endpoint, the expression is as follows: wherein is a binary operator that determines whether or not it is an end point of a spanning tree branch; According to the operator A mathematical model is established for solving the minimum number of endpoints, i.e. the minimum number of branches.

4. The full coverage path planning method for outdoor scenarios according to claim 3, wherein, According to the operator The expression of the mathematical model established is as follows:

5. The full coverage path planning method for outdoor scenarios according to claim 1, wherein, the turn cost of any node of the spanning tree is represented as: wherein denotes the degree of a node i.e. the number of its adjacent nodes; There are five ways of connecting any two spanning tree branches: "L" shape connection, "n" shape connection, "T" shape connection, "h" shape connection and "H" shape connection; connecting any two branches will change the degree of the connection node and , thereby affecting the total steering cost of the spanning tree ; change amount The expression is:

6. An electronic device, comprising: the electronic device 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 of any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that, 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 of any one of claims 1 to 5.

Images