Method, medium and system for testing comprehensive performance of mowing robot

Abstract

The invention belongs to the technical field of robot testing, and particularly relates to a comprehensive performance testing method, medium and system for a mowing robot, and the method comprises the following steps: S1, obtaining basic information: obtaining complete information of a test site and model parameters of the mowing robot, and calculating a theoretical optimal path through a path planning algorithm; s2, performance test: inputting test site fuzzy information to the mowing robot and starting performance test timing; s3, obtaining result information: obtaining a motion track of the mowing robot during testing and lawn states before and after mowing; s4, data analysis: calculating the coincidence rate of the motion trail of the mowing robot and the optimal path, and calculating the mowing coverage rate and the repeated mowing rate; and S5, evaluation and scoring: performing comprehensive scoring based on the coincidence rate, the mowing quality, the timing result, the mowing coverage rate and the repeated mowing rate. Compared with the prior art, the invention provides a suitable method for testing the mowing robot.

Description

Technical Field

[0001] The present invention belongs to the technical field of robot testing, and in particular relates to a comprehensive performance testing method, medium and system for a lawn mowing robot. Background Art

[0002] The global private lawns and gardens ownership is about 250 million, and the annual lawn mowing and care time is nearly 8 months. With the increasing acceptance of smart home and automation devices by consumers, more and more household and commercial users choose lawn mowing robots to replace traditional manual mowing. At the same time, the continuous emergence of new technologies such as AI perception, positioning, and lithium battery BMS, and the reduction of production costs, have greatly improved the performance of lawn mowing robots and broadened the market prospects. It is expected that the global market size of the lawn mowing robot industry will rise to about $2.3 billion in 2025 (from the "2024-2025 China Lawn Mowing Robot Industry Development Monitoring and Investment Strategy Research Report").

[0003] Regarding the evaluation system of the comprehensive performance of lawn mowing robots, it is the focus of major manufacturers and also a pain point where they are willing to invest costs. Under the existing technical conditions, how to scientifically and efficiently test, record data, and analyze data can not only help manufacturers optimize product performance but also provide reliable reference information for consumers.

[0004] However, there is currently a lack of a method for comprehensively testing the performance of lawn mowing robots. Therefore, it is necessary to propose a comprehensive performance testing method suitable for lawn mowing robots based on the test items of lawn mowing robots to comprehensively reflect the comprehensive performance of lawn mowing robots. Summary of the Invention

[0005] The purpose of the present invention is to provide a comprehensive performance testing method, medium and system for a lawn mowing robot to solve at least one of the above problems. The inventors of this application found that there is currently a lack of a method for comprehensively testing the performance of lawn mowing robots. Based on this, the inventors proposed a comprehensive performance testing method for lawn mowing robots to provide a suitable method for testing lawn mowing robots.

[0006] The purpose of the present invention is achieved through the following technical solutions:

[0007] The first aspect of the present invention discloses a comprehensive performance testing method for a lawn mowing robot, including the following steps:

[0008] S1: Obtain basic information: Obtain the complete information of the test site and the model parameters of the lawn mowing robot, and calculate the theoretical optimal path through the path planning algorithm;

[0009] S2: Performance testing: Input the fuzzy information of the test site into the lawn mowing robot and start the performance testing timing;

[0010] S3: Obtain result information: Obtain the motion trajectory during the lawn mowing robot test and the lawn state before and after mowing.

[0011] S4: Data analysis: Calculate the coincidence rate between the motion trajectory of the lawn mowing robot and the optimal path, and calculate the mowing coverage rate and the repeated mowing rate.

[0012] S5: Evaluation and scoring: Conduct a comprehensive score based on the coincidence rate, mowing quality, timing result, mowing coverage rate, and repeated mowing rate.

[0013] Preferably, the path planning algorithms include Dijkstra algorithm, topological map method, grid method, A* algorithm, RRT algorithm, D* algorithm, artificial potential field method, dynamic window method, and full coverage path planning algorithm.

[0014] Preferably, the fuzzy information of the test site is obtained by fuzzy processing of the complete information of the test site.

[0015] Preferably, the fuzzy processing includes Gaussian blur, mean filtering, median filtering, bilateral filtering, Kawase blur, and double blur.

[0016] Preferably, the timing is segmented timing, including initial path planning timing and mowing timing.

[0017] Preferably, the motion trajectory is obtained by a motion trajectory extraction algorithm, and the motion trajectory extraction algorithm includes feature point matching method, optical flow method, exclusive block matching method, region matching method, mean shift algorithm, and continuous adaptive mean shift algorithm.

[0018] Preferably, movable obstacles are also arranged in the test site;

[0019] In step S3, it also includes obtaining the positions of the movable obstacles during the test;

[0020] In step S4, it also includes calculating the total change distance of the movable obstacles;

[0021] In step S5, a comprehensive score is conducted based on the coincidence rate, mowing quality, timing result, total change distance of the movable obstacles, mowing coverage rate, and repeated mowing rate.

[0022] Preferably, calculating the total change distance of the movable obstacles is achieved through the following steps: First, identify the movable obstacles in the test site through an object detection algorithm and obtain the coordinate information of the movable obstacles in the test site; then, track the movable obstacles through a tracking algorithm; finally, calculate the position change of the movable obstacles during the performance test and accumulate to obtain the total change distance.

[0023] The second aspect of the present invention discloses a computer-readable storage medium, where the computer-readable storage medium includes a stored program. When the program runs, it controls the device where the computer-readable storage medium is located to execute any one of the above-mentioned test methods.

[0024] The third aspect of the present invention discloses a comprehensive performance test system for a lawn mowing robot, including: one or more processors, a memory, and one or more programs. Among them, the one or more programs are stored in the memory and are configured to be executed by the one or more processors. The one or more programs include those for executing any one of the above-mentioned test methods.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] In the method of the present invention, by inputting fuzzy information to the lawn mowing robot, it is required to optimize and adjust the path planning in real time based on incomplete initial conditions and the data obtained by sensors during the test process, which can better reflect the complex environmental conditions during the actual use of the lawn mowing robot. At the same time, by comparing the final movement trajectory of the lawn mowing robot with the theoretically optimal path calculated through complete information, its navigation planning and real-time optimization capabilities under complex environmental conditions can be tested. Further combining the timing results, the computing performance of the lawn mowing robot can be initially obtained.

[0027] In the method of the present invention, the lawn mowing performance, such as the lawn mowing quality, lawn mowing coverage rate, and repeated mowing rate, is mainly evaluated, which can comprehensively reflect the lawn mowing performance of the lawn mowing robot under normal use conditions.

[0028] In the method of the present invention, a scoring system is also adopted for comprehensive evaluation in combination with various data indicators, which can provide a more intuitive and easy-to-understand evaluation conclusion for users. Specific embodiments

[0029] The present invention will be described in detail below in combination with specific embodiments, but it is by no means a limitation to the present invention.

[0030] Embodiment

[0031] A comprehensive performance test method for a lawn mowing robot includes the following steps:

[0032] S1: Obtain basic information: Obtain the complete information of the test site and the model parameters of the lawn mowing robot, and calculate the theoretically optimal path through the path planning algorithm;

[0033] S2: Performance test: Input fuzzy information of the test site to the lawn mowing robot and start timing for the performance test;

[0034] S3: Obtain result information: Obtain the motion trajectory of the lawn mowing robot during the test and the lawn state before and after mowing.

[0035] S4: Data analysis: Calculate the coincidence rate between the motion trajectory of the lawn mowing robot and the optimal path, and calculate the mowing coverage rate and the repeated mowing rate.

[0036] S5: Evaluation and scoring: Conduct a comprehensive score based on the coincidence rate, mowing quality, timing result, mowing coverage rate, and repeated mowing rate.

[0037] More specifically, in this embodiment:

[0038] A comprehensive performance test method for a lawn mowing robot includes the following steps:

[0039] S1: Obtain basic information: Obtain the complete information of the test site and the model parameters of the lawn mowing robot, and calculate the theoretical optimal path through the path planning algorithm.

[0040] S2: Performance test: Input the fuzzy information of the test site to the lawn mowing robot and start timing the performance test.

[0041] S3: Obtain result information: Obtain the motion trajectory of the lawn mowing robot during the test, the positions of the movable obstacles, and the lawn state before and after mowing.

[0042] S4: Data analysis: Calculate the coincidence rate between the motion trajectory of the lawn mowing robot and the optimal path, calculate the total change distance of the movable obstacles, calculate the mowing coverage rate and the repeated mowing rate.

[0043] S5: Evaluation and scoring: Conduct a comprehensive score based on the coincidence rate, mowing quality, timing result, the total change distance of the movable obstacles, the mowing coverage rate, and the repeated mowing rate.

[0044] Among them:

[0045] The test site is enclosed by a fence.

[0046] The complete information of the test site includes the size of the test site (length, width), starting position, grass species used for the lawn, terrain (such as flat ground, ramp, pothole, etc.) and environmental information (such as soil humidity, hardness, weather conditions, light conditions), the setting positions of obstacles (including fixed and movable obstacles) and the distances between obstacles and other conventionally known information.

[0047] The model parameters of the lawn mowing robot are from the theoretical parameters provided by the test requester, including the size information of the lawn mowing robot, mowing width, speed and other information.

[0048] Path planning algorithms include Dijkstra algorithm, topological map method, grid method, A* algorithm, RRT algorithm, D* algorithm, artificial potential field method, dynamic window method, and full coverage path planning algorithm. An optimal path is selected from the results obtained by each algorithm. The iteration condition for this optimal path can be set as: the shortest length path when the mowing coverage rate reaches over 99%, or the shortest time-consuming path when the mowing coverage rate reaches over 99%.

[0049] Fuzzy information of the test site (in the way of picture input) is obtained by fuzzy processing of the complete information of the test site. Among them, the fuzzy processing can specifically adopt at least one of Gaussian blur, mean filtering, median filtering, bilateral filtering, Kawase blur, and double blur, so that the mowing robot can only obtain the approximate size information of the test site, the starting position, and the approximate position of obstacles.

[0050] Performance test timing usually includes two periods of timing. The first period is the time-consuming for the mowing robot to perform preliminary path planning processing after the fuzzy information is input, and the second period is the time-consuming for the mowing robot to start mowing.

[0051] The motion trajectory is obtained by extracting from the recorded video during the test process, specifically through a motion trajectory extraction algorithm. This motion trajectory extraction algorithm can be one of feature point matching method, optical flow method, exclusive block matching method, region matching method, mean shift algorithm, and continuous adaptive mean shift algorithm. By calculating the coincidence rate through comparing the motion trajectory with the theoretical optimal path, the positioning and navigation ability of the mowing robot can be initially judged.

[0052] The total change distance of the movable obstacle, similar to the motion trajectory, the main difference is that it is obtained by identifying and accumulating the position and total motion distance of the movable obstacle in the recorded video. Specifically, first, the target detection algorithm is used to identify the movable obstacles in the test site and obtain the coordinate information of the movable obstacles in the test site; then the tracking algorithm is used to track the movable obstacles; finally, the position change of the movable obstacles during the performance test is calculated and accumulated to obtain the total change distance; among them, the target detection algorithm can adopt YOLO or SSD, and the tracking algorithm can adopt KLT tracker or MeanShift tracker; the coordinate information of the movable obstacles can be obtained based on the coordinate system established for the test site.

[0053] The mowing coverage rate, which is the ratio of the mowed area to the theoretical mowing area after the mowing robot completes the mowing performance test; the repeated mowing rate, which is the ratio of the area of the repeated mowing region to the theoretical mowing area after the mowing robot completes the mowing performance test. Among them, the mowed area can be obtained through the analysis of the test result image obtained after the mowing test; the area of the repeated mowing region can be obtained through the video analysis during the mowing test (which needs to be calculated in combination with the movement trajectory of the mowing robot and its mowing width).

[0054] The mowing quality, including the uniformity and neatness of the lawn after mowing, can be obtained through the analysis of the test result image obtained after the mowing test.

[0055] When evaluating and scoring the performance of the mowing robot, a scoring system can be adopted and the total score can be calculated; different evaluation grades can be assigned according to the total score. Among them, the coincidence rate, mowing quality, timing result, total change distance of movable obstacles, mowing coverage rate and repeated mowing rate are quantized in different grades according to different test conditions and calculation results, and the quantization value of the test site environment is used as the coefficient for the quantization summation of the corresponding test performance results; based on this, the comprehensive performance of the mowing robot in different test environments can be obtained. The above quantization values and grading methods are established and adjusted according to existing standards, legal regulations and industry consensus.

[0056] The above description of the embodiments is to enable those of ordinary skill in the art to understand and use the invention. It is obvious that those skilled in the art can easily make various modifications to these embodiments and apply the general principles described herein to other embodiments without creative labor. Therefore, the present invention is not limited to the above embodiments, and the improvements and modifications made by those skilled in the art without departing from the scope of the present invention should be within the protection scope of the present invention.

Claims

1. A comprehensive performance testing method for a lawn mowing robot, characterized in that: The steps include: S1: Obtain basic information: Obtain complete information about the test site and model parameters of the mowing robot, and calculate the theoretical optimal path through the path planning algorithm; S2: Performance test: input the fuzzy information of the test site to the mowing robot and start the performance test timing; S3: Obtain result information: obtain the motion trajectory of the lawn mowing robot during the test and the lawn status before and after mowing; S4: Data analysis: Calculate the overlap rate between the motion trajectory of the mowing robot and the optimal path, calculate the mowing coverage rate and the repeated mowing rate; S5: Evaluation and scoring: Comprehensive scoring is based on overlap rate, mowing quality, timing results, mowing coverage rate and repeated mowing rate.

2. The comprehensive performance testing method of a lawn mowing robot according to claim 1, characterized in that: The path planning algorithms include Dijkstra algorithm, topological map method, grid method, A* algorithm, RRT algorithm, D* algorithm, artificial potential field method, dynamic window method and full coverage path planning algorithm.

3. The comprehensive performance testing method of a lawn mowing robot according to claim 1, characterized in that: The fuzzy information of the test site is obtained by fuzzy processing the complete information of the test site.

4. The comprehensive performance testing method of a lawn mowing robot according to claim 3, characterized in that: The blur processing includes Gaussian blur, mean filtering, median filtering, bilateral filtering, Kawase blur and double blur.

5. The comprehensive performance testing method of a lawn mowing robot according to claim 1, characterized in that: The timing is segmented timing, including initial path planning timing and mowing timing.

6. The comprehensive performance testing method of a lawn mowing robot according to claim 1, characterized in that: The motion trajectory is obtained through a motion trajectory extraction algorithm, which includes a feature point matching method, an optical flow method, an exclusive block matching method, a region matching method, a mean shift algorithm, and a continuous adaptive mean shift algorithm.

7. The comprehensive performance testing method of a lawn mowing robot according to claim 1, characterized in that: The test site is also provided with movable obstacles; Step S3 also includes obtaining the position of the movable obstacle during the test; Step S4 also includes calculating the total change distance of the movable obstacle; In step S5, a comprehensive score is given based on the overlap rate, mowing quality, timing result, total change distance of movable obstacles, mowing coverage rate and repeated mowing rate.

8. The comprehensive performance testing method of a lawn mowing robot according to claim 7, characterized in that: The calculation of the total change distance of the movable obstacle is achieved by the following steps: first, the movable obstacle in the test site is identified by the target detection algorithm and the coordinate information of the movable obstacle in the test site is obtained; then, the movable obstacle is tracked by the tracking algorithm; Finally, the position change of the movable obstacle during the performance test is calculated and the total change distance is accumulated.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the test method according to any one of claims 1 to 8.

10. A comprehensive performance test system for a lawn mowing robot, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the test method described in any one of claims 1 to 8.