Out-of-bounds prevention control method for robotic lawn mower

Abstract

An out-of-bounds prevention control method for a robotic lawn mower. The method comprises: equipping a robotic lawn mower with a sensor assembly to perform boundary line and obstacle detection, wherein the sensor assembly is composed of a boundary line sensor and an obstacle sensor (S1); implementing sub-region division, performing boundary segmentation and operating node setting on sub-regions, accurately measuring operating node coordinates, and constructing quadrilateral cells (S2); on the basis of terrain and obstacle conditions, adjusting the specifications of all the quadrilateral cells, so as to obtain minimized local meshes (S3); using a formula to perform quality optimization on the minimized local meshes, so as to acquire a target region (S4); and on the basis of the target region, analyzing a travelling path of the robotic lawn mower, so as to acquire an abnormal signal, and taking into account the abnormal signal, further determining whether the robotic lawn mower is out of bounds (S5).

Description

Mowing robot out-of-bound prevention control method TECHNICAL FIELD

[0001] The present application relates to the field of mowing robot algorithm analysis technology, especially to a mowing robot out-of-bound prevention control method. BACKGROUND

[0002] Algorithm analysis technology mainly focuses on time and space complexity, stability, correctness, etc., and improves execution efficiency and resource utilization through optimization analysis. Algorithm analysis technology plays a crucial role in mowing robots, not only improving the intelligence and automation level of mowing robots, but also significantly improving mowing efficiency and mowing effect.

[0003] With the continuous development of technology, algorithm analysis technology will continue to provide strong support for the progress and development of mowing robots in the application of mowing robots. However, there are indeed some defects in the current mowing robots in preventing out-of-bound, including the following aspects: insufficient positioning accuracy, relying only on a single sensor to prevent out-of-bound; poor environmental adaptability, the path planning and obstacle avoidance algorithm of the mowing robot may not fully adapt to the complex environment, and may cross the boundary; algorithm optimization is insufficient, some mowing robot algorithms may be relatively simple or outdated, and cannot well cope with complex environments and boundary changes, and may not be able to respond in time when encountering sudden obstacles, resulting in collision with obstacles or crossing the boundary. SUMMARY

[0004] The purpose of the present application is to solve the problems in the background art, and a mowing robot out-of-bound prevention control method is proposed.

[0005] In order to achieve the above purpose, the present application adopts the following technical scheme:

[0006] The mowing robot out-of-bound prevention control method comprises:

[0007] Step 1, provide a sensor assembly for the mowing robot to detect boundary lines and obstacles; wherein the sensor assembly is composed of a boundary line sensor and an obstacle sensor;

[0008] Step 2, implement sub-region division, boundary segmentation and working node setting for the sub-region, accurately measure the working node coordinates, and construct a quadrilateral single grid;

[0009] Step 3, based on the terrain and obstacle conditions, adjust the specifications of all quadrilateral single grids to obtain the minimum local grid; wherein the specifications of the quadrilateral single grid include its shape and size, and the specification adjustment includes the variation of the quadrilateral shape in the minimum local grid, i.e. making all quadrilaterals similar in shape;

[0010] Step four, quality optimization of the minimized local grid is carried out by using the formula, and the target area is obtained;

[0011] Step five, the driving path of the mowing robot is analyzed according to the target area, and an abnormal signal is obtained, and whether the mowing robot is out of the boundary is further judged combined with the abnormal signal.

[0012] It should be noted that the application object of the mowing robot out-of-boundary control method proposed by the application can be widely set as the monitoring and analysis of outdoor areas such as gardens, lawns, farmlands and other areas that need automatic management. It can be used to monitor whether the working state of the mowing robot is normal, including its moving path and obstacle avoidance capability. Specifically, it can be comprehensively monitored and accurately analyzed by sensor technology and algorithm analysis, real-time position of the mowing robot, surrounding environment, obstacle distance and preset boundary line, etc. The position of the mowing robot is tracked, and it is detected whether the robot is close to or crosses the preset boundary. The boundary marker or obstacle is identified, and it is intelligently judged whether the mowing robot is about to or has been out of the boundary. Once it is found that the mowing robot has the risk of out-of-boundary or has actually out-of-boundary, measures are taken immediately according to the early warning prompt, so as to intervene in time and ensure the safety and efficiency of the mowing work.

[0013] Further, the process of equipping the mowing robot with a sensor assembly for boundary line and obstacle detection includes:

[0014] Select the type of boundary line sensor according to the working area environment, i.e. electromagnetic line sensor, visible light or infrared boundary line sensor: for outdoor environment, use electromagnetic line sensor buried in the ground; for indoor or temporary arrangement scene, use visible light or infrared boundary line sensor on the ground;

[0015] Arrange the boundary line on the boundary of the working area, and determine the range of the mowing robot working area by monitoring the boundary line on the ground or the boundary line buried in the ground: for the electromagnetic line sensor buried in the ground, dig a trench and lay electromagnetic line; for the boundary line on the ground, directly arrange the cable or sticker with sensor on the boundary;

[0016] Calibrate the boundary line sensor, including setting the sensitivity of the sensor and adjusting the detection range of the sensor;

[0017] According to the type and characteristics of the obstacles appearing in the working area, select the type of obstacle sensor, i.e. ultrasonic or infrared sensor; among them, the ultrasonic sensor is used to detect most types of obstacles, and the infrared sensor is used to detect transparent or highly reflective obstacles;

[0018] Install ultrasonic or infrared sensors in all potential obstacle areas around the mowing robot;

[0019] Set the detection distance, detection angle, and sensitivity of the obstacle sensor to obtain the obstacle avoidance capability;

[0020] When the mowing robot is working, the sensor assembly detects the boundary line position and surrounding obstacles in real time, and the sensor assembly immediately transmits the detection results to the control module of the mowing robot; the control module determines whether the moving path of the mowing robot needs to be adjusted or other obstacle avoidance measures need to be taken according to the detection results;

[0021] It can be understood that in the use of the mowing robot, selecting the most suitable sensor of the sensor assembly to assist in detecting the boundary and obstacles is a decision-making process that comprehensively considers technical performance, cost-effectiveness, environmental adaptability, and user needs, and needs to find the best balance point between multiple factors to ensure that the mowing robot can meet functional requirements and be economically feasible.

[0022] Further, the process of implementing sub-area division, boundary segmentation, and work node setting for the sub-area, accurately measuring the work node coordinates, and constructing four-sided single grids includes:

[0023] Based on the characteristics of the work area, sub-area division is implemented, and the boundary lines of the sub-areas are evenly or regularly selected on the boundary;

[0024] According to the characteristics of the terrain environment, the boundary lines of the sub-areas are divided into straight line segments or curve segments;

[0025] Work nodes are set in each sub-area, high-precision measurement tools or equipment are used to accurately obtain the coordinate information of each work node, and special processing or marking is performed on the work nodes at special positions on the boundary line, i.e., positioning the key points of the mowing robot out of the boundary;

[0026] Four adjacent work nodes are taken as vertices to construct four-sided single grids.

[0027] Further, based on the terrain and obstacle conditions, the specifications of all four-sided single grids are adjusted to obtain the minimum local grid, and the process includes:

[0028] Step Z1, combined with the set vertex coordinate position, all point sets in the two-dimensional space are extracted ;

[0029] Step Z2, set the center coordinate as ,

[0030] wherein,

[0031] ;

[0032] Step Z3, the compactness is measured by the sum of squares of distances from the point set to the center:

[0033] ;

[0034] where C represents the compactness target value; i represents the number of all points, i = 1, 2, …, n; n is a positive integer;

[0035] Step Z4, the length of the four edges of the quadrilateral single cell h is obtained, respectively ;

[0036] Step Z5, the average edge length of the quadrilateral single cell h is defined as:

[0037] ;

[0038] Step Z6, uniformity is achieved by minimizing the variance of all quadrilateral average edge lengths:

[0039] ;

[0040] where U represents the uniformity target value; N represents the number of quadrilateral single cells; h represents the number of quadrilateral single cells; j represents the number of the four edges of the quadrilateral single cell, j = 1, 2, 3, 4; represents the weighted average value of all quadrilateral edge lengths; represents the weight of the quadrilateral edge length;

[0041] Step Z7, a small moving vector is defined for updating the position of point i;

[0042] Step Z8, the position of each point is adjusted through the iterative formula, and the size of the quadrilateral single cell is adjusted:

[0043] ;

[0044] where and represent the original coordinates of the i-th point in the point set P; and represent the new coordinates of the i-th point calculated through the iterative formula; represents the learning rate for controlling the step size of iteration; and represent the partial derivatives of uniformity U with respect to the a and b coordinates of point i; and represent the partial derivatives of compactness C with respect to the a and b coordinates of point i;

[0045] Step Z9, repeat step Z8 to update the coordinates of each point until the stopping criterion is met;

[0046] It can be understood that the stopping criterion can be that the maximum number of iterations is reached, the change of uniformity and compactness is less than a certain threshold, etc., and each iteration aims to make the point set closer to the ideal uniform and compact distribution;

[0047] Step Z10, based on the new coordinates after iteration, determines the edge length and angle of the quadrilateral single grid, and integrates to form a minimized local grid.

[0048] Further, the process of obtaining the target area includes:

[0049] The penalty term is introduced to optimize the quality of the minimized local grid:

[0050] ;

[0051] In the formula, GEI represents the grid quality optimization index; The introduced penalty term is an index for measuring the complexity of the grid boundary, which is used to punish the grid with irregular shape or complex boundary; and The maximum and minimum values of the edge length of the quadrilateral single grid after iterative calculation are used to measure the difference of the edge length; k represents the edge length difference adjustment factor, which is used to adjust the contribution degree of the edge length difference to the shape regularity; v represents the angle difference adjustment factor, which is used to adjust the contribution degree of the angle difference to the shape regularity; The internal angle of the quadrilateral single grid after iterative calculation is represented by The internal angle of the ideal shape is represented by The sum of all internal angles is represented by

[0052] The multiple quality-optimized minimized local grids are merged into a large grid set, and the integrated regional grid completely represents the target area.

[0053] Further, the process of obtaining the abnormal signal based on the analysis of the driving path of the mowing robot in the target area includes:

[0054] The data of the target area is sorted into a format that can be automatically recognized by the mowing robot.

[0055] After the mowing robot receives the optimized grid data, it is imported into the control module.

[0056] The target area G is determined, G=ST(x,y) represents the state of the (x,y)th quadrilateral single grid; wherein x and y represent the indices of rows and columns; the value of ST(x,y) is: 0 represents an empty land, i.e. the mowing robot can normally pass through; 1 represents an obstacle; 2 represents mowed grass;

[0057] The target area G is preprocessed to identify all the four-sided cells where the boundaries and obstacles are located, which are regarded as the unreachable places for the mowing robot;

[0058] The target area G is updated, and the four-sided cells with the value of 1 for ST(x, y) are automatically skipped;

[0059] The starting point Q and the target point E are set, which represent the initial position and the end point that needs to be reached for the mowing robot, respectively;

[0060] The shortest path from Q to E is calculated by the function f(g) = t(g) + q(g); where t(g) represents the actual cost from the starting point to the current four-sided cell g, and q(g) represents the estimated cost from the current four-sided cell to the target point;

[0061] It can be understood that q(g) is a heuristic function that is adjusted according to the unevenness of the terrain, and when a four-sided cell is far away from the nearest obstacle or the terrain is relatively flat, the value of q(g) is lower;

[0062] The minimum cost path from Q to each four-sided cell is calculated by the function ; where d(g) represents the minimum cost from the starting point to the current four-sided cell g, e(m, g) represents the moving cost from the four-sided cell m to g, M(g) represents the neighbor set of g, and d(m) represents the moving cost from the starting point to the four-sided cell m;

[0063] The target path π(g) = f(g) + d(g) for the mowing robot is generated by combining the shortest path and the minimum cost path;

[0064] The actual path of the mowing robot is compared and analyzed with the target path: the deviation ratio is calculated using the formula PL = (Dis / TL) x 100%, where PL represents the deviation ratio, Dis represents the total deviation distance between the actual path and the target path of the mowing robot, and TL represents the total length of the target path; the deviation threshold PL0 is defined, if PL is not less than PL0, an abnormal signal is generated, indicating that the actual path of the mowing robot deviates seriously from the target path; if PL is less than PL0, a normal signal is generated, indicating that the actual path of the mowing robot is close to the target path.

[0065] Further, the process of further judging whether the mowing robot is out of the boundary based on the abnormal signal includes:

[0066] For the abnormal signal, the current position D and the next position F of the mowing robot are captured using a high-precision positioning system;

[0067] The information of the known boundary four-sided cells is extracted: the coordinates of the four adjacent vertices of the boundary four-sided cells are distinguished: ​

[0068] In the formula, represents the coordinates of the top right vertex of the boundary quadrilateral single cell; represents the coordinates of the bottom right vertex of the boundary quadrilateral single cell; represents the coordinates of the top left vertex of the boundary quadrilateral single cell; represents the coordinates of the bottom left vertex of the boundary quadrilateral single cell;

[0069] A safety distance L is preset for the operation of the mowing robot, and the following is obtained:

[0070] ;

[0071] In the formula, represents the minimum distance from point F to line segment 13, represents the minimum distance from point F to line segment 14, represents the minimum distance from point F to line segment 24, represents the minimum distance from point F to line segment 23, wherein, represents the distance from point F to the line segment;

[0072] According to the early warning function B, the following is obtained:

[0073] ;

[0074] When the minimum distance from point F to any line segment is less than the safety distance L, B(D, F) = 1, indicating that the robot is approaching the boundary and an early warning signal is issued; otherwise, B(D, F) = 0, indicating that the robot is in a normal operating area and an inquiry signal is issued, and the out-of-boundary situation of the mowing robot is further detected.

[0075] Based on the early warning signal, the out-of-boundary reason of the mowing robot is investigated.

[0076] It can be understood that the investigation can be carried out in the following aspects: checking the robot settings: checking the working range, motion trajectory and other settings of the robot to confirm whether the robot is out of boundary due to improper settings; analyzing sensor data: analyzing the sensor components equipped by the robot to find the possible reasons causing the robot to be out of boundary; checking software programs: checking the software programs for controlling the robot to determine whether there are logical errors or program vulnerabilities causing abnormal behavior of the robot.

[0077] Compared with the prior art, the mowing robot out-of-boundary prevention control method provided by the application has the following advantages:

[0078] 1. This invention utilizes sensor components to detect boundary lines and obstacles, enabling timely detection and response, reducing collisions between the robot and boundaries or obstacles, and protecting the safety of the robot itself and its surrounding environment; by dividing a large working area into multiple smaller sub-areas, it facilitates management and optimization of the mowing path; by segmenting the boundaries of the sub-areas and setting working nodes, it provides a clear working area for the mowing robot, accurately measures the coordinates of the working nodes, ensures the accuracy of path planning, reduces errors, and constructs quadrilateral cells as basic working units, facilitating subsequent grid optimization and path analysis;

[0079] 2. This invention adjusts the specifications of all quadrilateral cells according to terrain and obstacle conditions to obtain a minimized local grid. The specifications of the quadrilateral cells are flexibly adjusted based on terrain undulations and obstacle distribution, making the grid layout more closely match the actual working environment, reducing unnecessary path movement, and improving mowing efficiency. The minimized local grid is then optimized using a formula to obtain the target area, ensuring that path planning is both efficient and reasonable, reducing repetitive work and idle time, and helping the mowing robot achieve precise operation and improve work quality.

[0080] 3. This invention monitors the driving path of the lawnmower robot in real time to ensure that it operates within the predetermined area. Once an abnormal signal is detected, it immediately makes a boundary judgment. By responding promptly to the abnormal signal, it quickly determines whether the lawnmower robot has crossed the boundary and takes corresponding measures to prevent further boundary crossing.

[0081] In summary, this invention can monitor and analyze the operating status of lawnmower robots working in outdoor areas according to actual conditions, ensuring that they operate within the predetermined area and avoiding damage caused by crossing boundaries. By using sensor technology and algorithm analysis to intelligently determine whether the lawnmower robot is about to or has already crossed the boundary, it achieves efficient and precise control to prevent lawnmower robots from crossing boundaries, ensuring the normal implementation of subsequent methods for preventing lawnmower robots from crossing boundaries. Attached Figure Description

[0082] Figure 1 is a flowchart of the method for preventing the lawnmower robot from going out of bounds proposed in this invention. Detailed Implementation

[0083] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0084] Referring to Figure 1, the method for preventing the lawnmower robot from going out of bounds includes:

[0085] Step one, provide the lawn mower robot with a sensor assembly for boundary line and obstacle detection; wherein the sensor assembly is composed of a boundary line sensor and an obstacle sensor;

[0086] Step two, implement sub-area division, boundary segmentation and working node setting for the sub-area, accurately measure the working node coordinates, and construct a quadrilateral single grid;

[0087] Step three, based on the terrain and obstacle situation, adjust the specifications of all quadrilateral single grids, including their shape and size, to obtain the minimum local grid; wherein the specification adjustment includes the variation of the quadrilateral shape in the minimum local grid, that is, to make all quadrilaterals similar in shape;

[0088] Step four, use the formula to optimize the quality of the minimum local grid to obtain the target area;

[0089] Step five, analyze the driving path of the lawn mower robot according to the target area, obtain abnormal signals, and further judge whether the lawn mower robot is out of the boundary based on the abnormal signals.

[0090] It should be noted that the application object of the lawn mower robot out-of-boundary prevention control method proposed by the present application can be widely set as the monitoring and analysis of outdoor areas such as gardens, lawns, farmland and other areas that need automatic management. It can be used to monitor whether the working state of the lawn mower robot is normal, including its moving path and obstacle avoidance ability. Specifically, it can be analyzed by sensor technology and algorithm, and the real-time position of the lawn mower robot, the surrounding environment, the distance of the obstacle and the preset boundary line are comprehensively monitored and accurately analyzed to track the position of the lawn mower robot, detect whether the robot is close to or crosses the preset boundary, identify boundary markers or obstacles, intelligently judge whether the lawn mower robot is about to or has already gone out of the boundary, and take measures immediately according to the warning prompt to intervene in time to ensure the safety and efficiency of the mowing work.

[0091] Please refer to FIG. 1, the present application provides a lawn mower robot out-of-boundary prevention control method, the step one, for the lawn mower robot equipped with a sensor assembly for boundary line and obstacle detection, including:

[0092] Step 101, select the type of boundary line sensor according to the working area environment, that is, electromagnetic line sensor, visible light or infrared boundary line sensor: for outdoor environment, use the electromagnetic line sensor buried in the ground; for indoor or temporarily arranged scene, use visible light or infrared boundary line sensor on the ground;

[0093] Step 102, arranging boundary lines on the boundary of the working area to determine the range of the mowing robot working area by monitoring the boundary lines on the ground or buried boundary lines: for the buried electromagnetic line sensor, dig a trench underground and lay the electromagnetic line; for the boundary line on the ground, directly arrange the cable or sticker with sensors on the boundary;

[0094] Step 103, calibrate the boundary line sensor, including setting the sensitivity of the sensor and adjusting the detection range of the sensor;

[0095] Step 104, select the type of obstacle sensor according to the type and characteristics of the obstacles appearing in the working area, i.e. ultrasonic or infrared sensor; wherein the ultrasonic sensor is used to detect most types of obstacles, while the infrared sensor is used to detect transparent or highly reflective obstacles;

[0096] Step 105, install ultrasonic or infrared sensors in all potential obstacle areas around the mowing robot;

[0097] Step 106, set the detection distance, detection angle, and sensitivity of the obstacle sensor to obtain the obstacle avoidance capability;

[0098] Step 107, when the mowing robot is working, the sensor assembly detects the boundary line position and the surrounding obstacles in real time, and the sensor assembly immediately transmits the detection results to the control module of the mowing robot; the control module determines whether to adjust the moving path of the mowing robot or take other obstacle avoidance measures according to the detection results;

[0099] In steps 101-107, in actual application process, comprehensive consideration budget limit, working environment characteristics, sensor performance and cost control, post-maintenance and upgrade, and technical maturity and reliability and other factors, select the most suitable sensor assembly for current project demand to assist the mowing robot to carry out boundary detection and obstacle avoidance, therefore, need to find the best balance point between multiple factors, to ensure that the mowing robot can meet the functional requirements, and can be economically feasible.

[0100] Referring to FIG. 1, the present application provides a method for preventing the mowing robot from going out of the boundary, and the step two, the implementation of sub-area division, the boundary segmentation of sub-area and the working node setting, the accurate measurement of working node coordinates, and the construction of four-edge single grid steps include:

[0101] Step 201, based on the characteristics of the working area, implement sub-area division, and evenly or regularly select the boundary lines of the sub-area on the boundary;

[0102] Step 202, divide the boundary lines of the sub-area into straight line segments or curve segments according to the characteristics of the terrain environment;

[0103] Step 203, set working nodes in each sub-region, accurately obtain coordinate information of each working node using high-precision measurement tools or equipment, and specially process or mark the working nodes at special positions on the boundary line, that is, locate the out-of-bound key points of the mowing robot;

[0104] Step 204, construct a quadrilateral single cell with four adjacent working nodes as vertices.

[0105] Referring to FIG. 1, the present application provides a control method for preventing the mowing robot from going out of bounds, and the step of adjusting the specifications of all the quadrilateral single cells based on the terrain and obstacle conditions in step three includes:

[0106] Step 301, combine the set vertex coordinate positions to extract all point sets in a two-dimensional space .

[0107] Step 302, set the center coordinate as ,

[0108] wherein .

[0109] Step 303, measure the compactness by the sum of squares of distances between the point sets and the center:

[0110] .

[0111] In the formula, C represents the compactness target value; i represents the number of all points, i=1, 2, …, n; n is a positive integer;

[0112] Step 304, obtain the lengths of the four edges of the quadrilateral single cell h, which are .

[0113] Step 305, define the average edge length of the quadrilateral single cell h as:

[0114] .

[0115] Step 306, realize uniformity by minimizing the variance of the average edge length of all quadrilaterals:

[0116] .

[0117] In the formula, U represents the uniformity target value; N represents the number of quadrilateral single cells; h represents the number of quadrilateral single cells; j represents the number of the four edges of the quadrilateral single cell, j=1, 2, 3, 4; represents the weighted average value of all quadrilateral edge lengths; represents the weight of the quadrilateral edge length;

[0118] Step 307, defining a small movement vector , for updating the position of point i;

[0119] Step 308, adjusting the position of each point through an iterative formula, and adjusting the size of the quadrilateral single grid:

[0120] ;

[0121] wherein, and represent the original coordinates of the i-th point in the point set P; and represent the new coordinates of the i-th point calculated through the iterative formula; represents the learning rate for controlling the step size of iteration; and represent the partial derivatives of uniformity U with respect to the a and b coordinates of point i; and represent the partial derivatives of compactness C with respect to the a and b coordinates of point i;

[0122] Step 309, repeating step 308 to update the coordinates of each point until the stopping criterion is met;

[0123] In step 309, the stopping criterion can be reaching the maximum number of iterations, the change of uniformity and compactness being less than a certain threshold, etc., and each iteration aims to make the point set closer to the ideal uniform and compact distribution;

[0124] Step 310, based on the new coordinates after iteration, determining the side length and angle of the quadrilateral single grid, and integrating to form the minimized local grid.

[0125] Referring to FIG. 1, the present application provides a method for preventing the mowing robot from going out of the boundary, and the step four includes:

[0126] Step 401, introducing a penalty term to optimize the quality of the minimized local grid:

[0127] ;

[0128] wherein, GEI represents the grid quality optimization index; is the introduced penalty term, which is an index for measuring the complexity of the grid boundary, and is used to punish the grid with irregular shape or complex boundary; and represents the maximum and minimum values of the side length of the quadrilateral single grid after iterative calculation, used to measure the difference of the side length; k represents the side length difference adjustment factor, used to adjust the contribution degree of the side length difference to the shape regularity; v represents the angle difference adjustment factor, used to adjust the contribution degree of the angle difference to the shape regularity; represents the internal angle of the quadrilateral single grid after iterative calculation; represents the internal angle of the ideal shape; represents the sum of all internal angles;

[0129] Step 402, a plurality of quality-optimized minimized local grids are merged into a large grid set, and the integrated regional grid completely represents the target region.

[0130] Referring to FIG. 1, the present application provides a control method for preventing the mowing robot from going out of the boundary, and the step of analyzing the driving path of the mowing robot according to the target region to obtain an abnormal signal, and further judging whether the mowing robot goes out of the boundary in combination with the abnormal signal comprises the following steps:

[0131] Step 501, data of the target region is arranged into a format that can be automatically recognized by the mowing robot;

[0132] Step 502, after the mowing robot receives the optimized grid data, the mowing robot imports the grid data into a control module;

[0133] Step 503, the target region G is determined, G=ST(x, y) represents the state of the (x, y)th quadrilateral single grid; wherein x and y represent the indices of rows and columns; the value of ST(x, y) is: 0 represents an empty land, i.e., the mowing robot can normally pass through; 1 represents an obstacle; and 2 represents a mowed land;

[0134] Step 504, the target region G is preprocessed, and all quadrilateral single grids where the boundaries and obstacles are located are identified as places where the mowing robot cannot reach;

[0135] Step 505, the target region G is updated, and the quadrilateral single grids whose values of ST(x, y) are 1 are automatically skipped;

[0136] Step 506, a starting point Q and a target point E are set, which respectively represent the initial position of the mowing robot and the end point that needs to be reached;

[0137] Step 507, the shortest path from Q to E is calculated through a function f(g)=t(g)+q(g); wherein t(g) represents the actual cost from the starting point to the current quadrilateral single grid g, and q(g) represents the estimated cost from the current quadrilateral single grid to the target point;

[0138] In step 507, q(g) is a heuristic function, which is adjusted according to the unevenness of the terrain, and when the nearest obstacle of a certain quadrilateral cell is far away or the terrain is relatively flat, the value of q(g) is low;

[0139] Step 508, calculate the minimum cost path Q to each quadrilateral cell; wherein d(g) represents the minimum cost from the starting point to the current quadrilateral cell g, e(m, g) represents the moving cost from the quadrilateral cell m to g, M(g) represents the neighbor set of g, and d(m) represents the moving cost from the starting point to the quadrilateral cell m;

[0140] Step 509, generate the target path of the mowing robot in combination with the shortest path and the minimum cost path, π(g) = f(g) + d(g);

[0141] Step 510, compare and analyze the actual path of the mowing robot with the target path: calculate the deviation ratio using the formula PL = (Dis / TL) x 100%, wherein PL represents the deviation ratio, Dis represents the total deviation distance between the actual path of the mowing robot and the target path, and TL represents the total length of the target path; define a deviation threshold PL0, if PL is not less than PL0, an abnormal signal is generated, indicating that the actual path of the mowing robot deviates from the target path seriously; if PL is less than PL0, a normal signal is generated, indicating that the actual path of the mowing robot is close to the target path;

[0142] Step 511, for the abnormal signal, use a high-precision positioning system to capture the current position D and the next position F of the mowing robot;

[0143] Step 512, extract the information of the known boundary quadrilateral cell: distinguish the coordinates of the four adjacent vertices of the boundary quadrilateral cell:

[0144] wherein, represents the coordinates of the top-right vertex of the boundary quadrilateral cell; represents the coordinates of the bottom-right vertex of the boundary quadrilateral cell; represents the coordinates of the top-left vertex of the boundary quadrilateral cell; represents the coordinates of the bottom-left vertex of the boundary quadrilateral cell;

[0145] Step 513, preset a safety distance L for the operation of the mowing robot, then:

[0146]

[0147] wherein, represents the minimum distance from the point F to the line segment 13, ​​​represents the minimum distance of point F to line segment 14, represents the minimum distance of point F to line segment 24, represents the minimum distance of point F to line segment 23, wherein, represents the distance of point F to line segment;

[0148] Step 514, according to the early warning function B to make a judgment, in order to take immediate measures to prevent the robot out of the border:

[0149] ;

[0150] When the minimum distance of point F to any line segment is less than the safety distance L, B(D, F) = 1, which indicates that the robot is approaching the boundary, and an early warning signal is issued; otherwise B(D, F) = 0, which indicates that the robot is in the normal operation area, and an inquiry signal is issued, and the further detection of the lawn mower out of the border is carried out;

[0151] Step 515, based on the early warning signal to investigate the reason of the lawn mower out of the border, improve the working efficiency and accuracy of the robot;

[0152] In step 515, the following aspects can be investigated: check the robot settings: check the working range, motion trajectory and other settings of the robot, and confirm whether the robot out of the border is caused by improper settings; analyze sensor data: analyze the sensor components equipped by the robot, and find out the possible reasons that cause the robot out of the border; check the software program: check the software program that controls the robot, whether there is logical error or program bug that causes the robot abnormal behavior.

[0153] In the embodiment of the present application, the lawn mowing robot is configured with boundary line sensors and obstacle sensors to perceive and identify the boundary line and obstacles, which is the basis for preventing the robot from going out of the boundary. The boundary line sensors help the robot identify and follow the preset boundary line, while the obstacle sensors ensure that the robot can avoid or take other measures in time when encountering obstacles. Through sub-area division and the construction of quadrilateral single grids, the robot can divide its working area into smaller and more manageable parts, effectively control the movement range and path of the robot, and reduce the possibility of repeated work and missed areas. At the same time, the grid size is adjusted according to the terrain and obstacle conditions to obtain the minimized local grid, which is an optimization of the preliminary division of the quadrilateral single grid to adapt to the actual terrain and obstacle conditions. By adjusting the size of the single grid, the robot can still work effectively when encountering complex terrain or a large number of obstacles, while reducing unnecessary path and energy consumption. Through quality optimization of the minimized local grid using the formula, the target area is obtained, improving the quality and efficiency of the grid, making the optimized grid more suitable for the robot's work requirements, and improving work efficiency and accuracy. By monitoring and analyzing the robot's driving path in real time based on the optimized grid and target area, measures are taken immediately to prevent the robot from going out of the boundary once abnormal signals (such as deviating from the preset path or crossing the boundary line) are detected. In summary, the present application relates to data processing, comprehensive analysis and optimization operation decision-making, solving the technical problem of poor out-of-boundary control effect caused by the lack of precise analysis of the environment adaptability and driving path of the lawn mowing robot in the existing scheme. In actual situations, more data and contextual information may be needed to make specific decisions and optimization schemes.

[0154] In addition, the formulas involved in the above are calculated by removing the dimension and taking the numerical value, which is obtained by collecting a large amount of data to simulate the closest real situation. The proportionality coefficient in the formula and the various preset thresholds in the analysis process are set by the person skilled in the art according to the actual situation or obtained by a large amount of data simulation. The size of the proportionality coefficient is a specific numerical value obtained by quantifying each parameter for subsequent comparison. The size of the proportionality coefficient depends on the amount of sample data and the preliminary processing coefficient set by the person skilled in the art for each group of sample data.

[0155] Each embodiment in the specification is described in a progressive manner, and the same or similar parts of each embodiment can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically based on the method embodiment, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

[0156] For convenience of description, the above apparatus is described by dividing into various units according to functions and is described separately. Of course, functions of the units can be implemented in one or more software and / or hardware in implementing the present application.

[0157] Those skilled in the art should understand that embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take a form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take a form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0158] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and a combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus generate means for implementing the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0159] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including instruction means, which implements the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0160] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide steps for implementing the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0161] Secondly, the present application discloses only structures related to the embodiments of the present disclosure in the accompanying drawings, and other structures can refer to general designs. In the case of no conflict, the same embodiments and different embodiments of the present application can be combined with each other.

[0162] Finally: the above is only the preferred embodiment of the present application, but the protection scope of the present application is not limited to this, any person skilled in the art in the technical range disclosed by the present application, according to the technical scheme of the present application and its inventive concept, equivalent replacement or change, should be covered in the protection scope of the present application.

Claims

1.A method for preventing a mowing robot from going out of a boundary, characterized in that: Step 1: providing a sensor assembly for the mowing robot to detect the boundary line and obstacles; wherein the sensor assembly is composed of a boundary line sensor and an obstacle sensor; Step 2: implementing sub-region division, boundary segmentation and working node setting for the sub-region, accurately measuring the working node coordinates, and constructing a quadrilateral single grid; Step 3: adjusting the specifications of all quadrilateral single grids based on the terrain and obstacle conditions to obtain a minimized local grid; Step 4: using a formula to optimize the quality of the minimized local grid to obtain a target region; Step 5: analyzing the travel path of the mowing robot according to the target region, obtaining an abnormal signal, and further determining whether the mowing robot goes out of the boundary based on the abnormal signal; wherein the process of Step 3 includes: Step Z1, in combination with the set vertex coordinate position of the quadrilateral single cell, all point sets in the two-dimensional space are extracted ； Step Z2, let the center coordinate be ， wherein ； Step Z3: measuring compactness by the sum of squares of distances between the point set and the center: ； wherein C represents the compactness target value; i represents the number of all points, i = 1, 2, …, n; n is a positive integer; Step Z4, get the length of the four edges of the quadrilateral single cell h, respectively ； Step Z5: defining the average side length of the quadrilateral single grid h as: ； Step Z6: realizing uniformity by minimizing the variance of all quadrilateral average side lengths: ； In the formula, U represents a uniformity target value; N represents the number of quadrilateral single cells; h represents the number of the quadrilateral single cell; j represents the number of the four edges of the quadrilateral single cell, j = 1, 2, 3, 4; representing a weighted average of all the quadrilateral side lengths; denotes the weight of the quadrilateral side length; Step Z7, define a small movement vector is used to update the position of point i; Step Z8: adjusting the position of each point through the iterative formula, and further adjusting the specifications of the quadrilateral single grid: ； In the formulae, and represents the original coordinates of the i-th point in the point set P; and denotes the new coordinates of the i-th point calculated by the iterative formula; denotes a learning rate for controlling the step size of the iteration; and denotes the partial derivative of the uniformity U with respect to the a and b coordinates of the point i; and denotes the partial derivative of compactness C with respect to the a and b coordinates of point i; Step Z9: repeating Step Z8 to update the coordinates of each point until the stopping criterion is met; Step Z10: based on the new coordinates after iteration, determining the quadrilateral single grid side length and angle, and integrating to form a minimized local grid; wherein the process of Step 4 includes: introducing a penalty term to optimize the quality of the minimized local grid: ； In the formula, GEI represents a grid quality optimization index. For the introduced penalty term, is an index to measure the complexity of the grid boundary, which is used to punish the grid with irregular shape or complex boundary; and represents the maximum and minimum values of the side length of the quadrilateral single cell after the iterative calculation, for measuring the difference of the side length; k represents a side length difference adjustment factor, for adjusting the contribution degree of the side length difference to the shape regularity; v represents an angle difference adjustment factor, for adjusting the contribution degree of the angle difference to the shape regularity; denotes the inner angle of the quadrangular cell after the iterative calculation; an inner angle representing an ideal shape; denotes the sum of all internal angles; combining multiple quality-optimized minimized local grids into a large grid set, and the integrated regional grid completely represents the target region. 2.The method of claim 1, wherein the method further comprises: determining whether the cutting robot is located in the cutting area; and controlling the cutting robot to move to the cutting area when the cutting robot is determined to be located outside the cutting area. In Step 1, the process of providing a sensor assembly for the mowing robot to detect the boundary line and obstacles includes: selecting a boundary line sensor type according to the working area environment; arranging boundary lines on the boundary of the working area to determine the range of the mowing robot working area by monitoring the boundary lines on the ground or buried in the ground; calibrating the boundary line sensor; selecting an obstacle sensor type according to the type and characteristics of obstacles appearing in the working area; installing obstacle sensors in all potential obstacle areas around the mowing robot to detect whether there are obstacles interfering with the operation of the mowing robot; setting the detection distance, detection angle and sensitivity of the obstacle sensor to obtain the obstacle avoidance capability of the obstacle sensor; when the mowing robot is working, the sensor assembly detects the boundary line position and the surrounding obstacles in real time, and transmits the detection results to the control module of the mowing robot. 3.The method of claim 1, wherein the method further comprises: determining whether the cutting robot is located in the cutting area; and controlling the cutting robot to move to the cutting area when the cutting robot is determined to be located outside the cutting area. The step two, the process of implementing sub-region division, boundary segmentation and work node setting for the sub-region, accurately measuring the work node coordinates, and constructing a quadrilateral single grid includes: Implementing sub-region division based on the characteristics of the work area, and evenly or regularly screening the boundary lines of the sub-region on the boundary; Dividing the boundary lines of the sub-region into straight line segments or curve segments according to the characteristics of the terrain environment; Setting a work node in each sub-region, using high-precision measuring tools or equipment to accurately obtain the coordinate information of each work node, and simultaneously performing special processing or marking on the work nodes at special positions on the boundary line, i.e. positioning the key points for the robot to exit the boundary; Constructing a quadrilateral single grid with four adjacent work nodes as the vertices. 4.The method of claim 1, wherein the method further comprises: determining whether the cutting robot is located in the cutting area; and controlling the cutting robot to move to the cutting area when the cutting robot is determined to be located outside the cutting area. The step five, the process of analyzing the driving path of the mowing robot according to the target area to obtain an abnormal signal includes: Organizing the data of the target area into a format that can be automatically recognized by the mowing robot; After the mowing robot receives the optimized grid data, importing it into the control module; Determining the target area G, G=ST(x, y) represents the state of the (x, y)th quadrilateral single grid; wherein x and y represent the indices of rows and columns; the value of ST(x, y) is: 0 represents an empty land, i.e. the mowing robot can normally pass through; 1 represents an obstacle; 2 represents a mowed area; Pretreating the target area G, identifying all quadrilateral single grids where the boundaries and obstacles are located, and regarding them as places that the mowing robot cannot reach; Updating the target area G, and automatically skipping the quadrilateral single grids with a value of 1 for ST(x, y); Setting a starting point Q and a target point E, representing the initial position and the end point that needs to be reached of the mowing robot, respectively; Calculating the shortest path from Q to E through the function f(g)=t(g)+q(g); wherein t(g) represents the actual cost from the starting point to the current quadrilateral single grid g, and q(g) represents the estimated cost from the current quadrilateral single grid to the target point; By function Calculating the minimum cost path from Q to each quadrilateral single grid; wherein d(g) represents the minimum cost from the starting point to the current quadrilateral single grid g, e(m, g) represents the moving cost from the quadrilateral single grid m to g, M(g) represents the neighbor set of g, and d(m) represents the moving cost from the starting point to the quadrilateral single grid m; Generating the target path π(g) of the mowing robot in combination with the shortest path and the minimum cost path f(g)+d(g); Comparing and analyzing the actual path of the mowing robot with the target path: calculating the deviation ratio using the formula PL=(Dis / TL)×100%, wherein PL represents the deviation ratio, Dis represents the total deviation distance between the actual path and the target path of the mowing robot, and TL represents the total length of the target path; defining a deviation threshold PL0, if PL is not less than PL0, an abnormal signal is generated, indicating that the actual path of the mowing robot deviates from the target path seriously; if PL is less than PL0, a normal signal is generated, indicating that the actual path of the mowing robot is close to the target path. 5.The method of claim 1, wherein the method further comprises: determining whether the cutting robot is located in the cutting area; and controlling the cutting robot to move to the cutting area when the cutting robot is determined to be located outside the cutting area. The step five, the process of further judging whether the mowing robot exits the boundary in combination with the abnormal signal includes: For abnormal signals, the current position D and the next position F of the mowing robot are captured using a high-precision positioning system; extracting information of the known boundary quadrangle single cell: distinguishing the four adjacent vertex coordinate positions of the boundary quadrangle single cell: ； In the formulae, represents the coordinates of the top right vertex of the boundary quadrilateral single cell; represents the coordinate of the right lower vertex of the boundary quadrangle single cell; represents the coordinate of the upper left vertex of the boundary quadrangle single cell; The coordinates of the lower-left vertex of the boundary quadrilateral single cell are represented as; A safety distance L for the mowing robot to run is preset, and thus: ； In the formulae, denotes the minimum distance of the point F to the line segment 13, denotes the minimum distance of the point F to the line segment 14, denotes the minimum distance of the point F to the line segment 24, denotes the minimum distance of the point F to the line segment 23, wherein The distance from the point F to the line segment is represented as; A judgment is made according to the early warning function B: ； When the minimum distance from the point F to any line segment is less than the safety distance L, B(D, F) = 1, which indicates that the robot is approaching the boundary and an early warning signal is sent; otherwise, B(D, F) = 0, which indicates that the robot is in a normal running area and an inquiry signal is sent, and the out-of-bound situation of the mowing robot is further detected; The out-of-bound reason of the mowing robot is investigated based on the early warning signal.

Images