A mine rescue robot path planning simulation test method and system

Abstract

The present invention discloses a mine rescue robot path planning simulation test method and system, comprising the following steps: step S1, based on the mine historical accident big data, multiple groups of mine historical accident scenes are obtained, and scene features are extracted for the mine historical accident scenes, and then the multiple groups of mine historical accidents are discretely analyzed based on the scene features to obtain rescue scene features; step S2, a rescue simulation scene is constructed according to the rescue scene features, and the accident rescue points and safety points of the mine rescue robot are marked in the rescue simulation scene; step S3, multiple groups of rescue planning paths are analyzed for safety and efficiency to obtain simulation test evaluation values ​​of each group of path planning algorithms. The present invention obtains balanced rescue scene features and extreme rescue scene features in the rescue scene features to construct balanced rescue simulation scenes and extreme rescue simulation scenes, so as to accurately meet the simulation test needs with a limited number of rescue simulation scenes and avoid invalid simulation.

Description

Technical Field

[0001] The present invention relates to the technical field of rescue simulation testing, and in particular to a mine rescue robot path planning simulation testing method and system. Background Art

[0002] Coal is the main energy source in my country, and its production and consumption account for 76% and 68% of the total primary energy production and consumption respectively. With the increase of mining depth, the impact of disasters such as ground stress, heat damage, and gas outburst on mine safety production is more prominent, especially the thermal power disaster accidents in mines, which are characterized by strong suddenness, rapid development of fire, difficulty in disaster relief, and easy to cause secondary injury accidents, resulting in a large number of casualties and resource losses. The rescue of thermal power disaster accidents in mines is related to the lives of miners in distress and rescue personnel, and the overall situation of national development and stability. However, after the disaster occurs, the connection between the mine and the underground is often cut off, making it impossible to effectively collect the specific disaster situation underground, which becomes the biggest obstacle to rescue. Once the rescue decision is wrong, it may cause casualties to rescue personnel and cause the accident to expand. A successful emergency rescue depends not only on the advanced level of rescue equipment, but also on the correct understanding and scientific analysis of various disaster information by the command decision makers, as well as the rapid and accurate judgment of the severity of the disaster and the safety of rescue, and scientific and decisive disaster relief decisions.

[0003] The prior art discloses a variety of robot rescue path planning methods and systems, which can obtain rescue paths with excellent performance in terms of accuracy, efficiency, cost, safety, etc. At that time, such rescue path planning methods only expressed rescue indicators in theory, and real-life simulation tests were still required to obtain practical results of the rescue paths. However, the existing simulation test methods can only build accident scenarios based on human experience, which are highly subjective and random, and ultimately lead to inaccurate simulation test results. Summary of the invention

[0004] The purpose of the present invention is to provide a mine rescue robot path planning simulation test method and system to solve the technical problem that the prior art uses accident scenes constructed based on human experience, which is highly subjective and random, and ultimately leads to inaccurate simulation test results.

[0005] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0006] A mine rescue robot path planning simulation test method comprises the following steps:

[0007] Step S1, obtaining multiple groups of historical mine accident scenes based on the big data of historical mine accidents, extracting scene features of the historical mine accident scenes, and then performing discrete analysis on the multiple groups of historical mine accidents based on the scene features to obtain rescue scene features;

[0008] Step S2: construct a rescue simulation scene according to the rescue scene characteristics, and mark the accident rescue points and safety points of the mine rescue robot in the rescue simulation scene;

[0009] Step S3, performing rescue path planning on multiple groups of path planning algorithms of the mine rescue robot based on the rescue simulation scenario to obtain multiple groups of rescue planning paths, performing safety and efficiency analysis on the multiple groups of rescue planning paths to obtain simulation test evaluation values ​​of each group of path planning algorithms, so as to implement scenario simulation testing on the path planning function of the mine rescue robot;

[0010] Step S4: Adaptively match multiple groups of rescue planning algorithms to rescue simulation scenarios based on simulation test evaluation values ​​to achieve multi-scenario rescue operations of mine rescue robots.

[0011] As a preferred solution of the present invention, the extraction of scene features of historical mine accident scenes includes:

[0012] In each historical accident scene of the mine, the geological structure characteristics, mine structure characteristics, accident cause characteristics, accident site personnel distribution characteristics, mine task progress characteristics, rescue channel structure characteristics, accident site meteorological characteristics, and accident site structure characteristics are extracted in turn;

[0013] Principal component analysis is performed on the geological structure characteristics, mine structure characteristics, accident cause characteristics, accident site personnel distribution characteristics, mine task progress characteristics, rescue channel structure characteristics, accident site meteorological characteristics and accident site structure characteristics to obtain the information contribution value of each feature item in turn;

[0014] All feature items whose information contribution values ​​are higher than the preset threshold are regarded as scene features.

[0015] As a preferred solution of the present invention, the discrete analysis of multiple groups of historical mine accidents based on scene characteristics to obtain rescue scene characteristics includes:

[0016] The expected value of the rescue scene feature is set, and the expected value of the rescue scene feature and the mine historical accident scene features of multiple groups of mine historical accidents are discretely quantified to obtain the scene feature discreteness. The calculation formula of the scene feature discreteness is:

[0017]

[0018] In the formula, σ X Characterized as scene feature discreteness, X i The vector form of the mine accident scene characteristics representing the i-th mine accident, X E The vector form of the expected value of the rescue scene features, |X i-X E | Characterized by X i and X E The Euclidean distance of , N represents the total number of historical accidents in the mine;

[0019] Constructing the solution constraint condition of the expected value of the rescue scene feature, the solution constraint condition is:

[0020] X E,j ∈[min(X (1,N),j ),max(X (1,N),j )];

[0021] In the formula, X E,j Characterized by X E The jth eigencomponent in X (1,N) The mine accident scene characteristics represented by the 1st to Nth mine accident history are represented by X (1,N) The jth characteristic component in is represented by min(X (1,N),j )、max(X (1,N),j ) are represented as (1,N),j A function for obtaining the minimum value of the j-th characteristic component and a function for obtaining the maximum value of the j-th characteristic component;

[0022] Based on the solution constraint condition, the discreteness of the scene feature is minimized to obtain the equilibrium expected value of the rescue scene feature, and based on the solution constraint condition, the discreteness of the scene feature is maximized to obtain the extreme expected value of the rescue scene feature, wherein the equilibrium expected value of the rescue scene feature is a balanced expression of the mine historical accident scene features of multiple mine historical accidents, and the extreme expected value of the rescue scene feature is an extreme expression of the mine historical accident scene features of multiple mine historical accidents.

[0023] As a preferred solution of the present invention, the rescue simulation scene is constructed according to the rescue scene characteristics, including:

[0024] A balanced rescue simulation scenario is constructed based on the balanced expected value of the rescue scenario characteristics, and an extreme rescue simulation scenario is constructed based on the extreme expected value of the rescue scenario characteristics.

[0025] As a preferred solution of the present invention, the accident rescue point and the safety point of the mine rescue robot are marked in the rescue simulation scene, including:

[0026] In the balanced rescue simulation scenario and the extreme rescue simulation scenario, the distribution points of the personnel at the accident scene are marked as accident rescue points, and in the balanced rescue simulation scenario and the extreme rescue simulation scenario, the exits of the rescue passages are marked as safety points.

[0027] As a preferred solution of the present invention, the method of performing rescue path planning based on a rescue simulation scenario using multiple groups of path planning algorithms of a mine rescue robot to obtain multiple groups of rescue planning paths includes:

[0028] Based on the balanced rescue simulation scenario, multiple groups of path planning algorithms of the mine rescue robot are sequentially used to plan rescue paths to obtain multiple groups of balanced rescue planning paths;

[0029] The multiple path planning algorithms of the mine rescue robot are used to plan rescue paths in sequence based on extreme rescue simulation scenarios to obtain multiple groups of extreme rescue planning paths.

[0030] As a preferred solution of the present invention, the safety and efficiency analysis of multiple groups of rescue planning paths to obtain a simulation test evaluation value of each group of path planning algorithms includes:

[0031] The number of accident rescue points included in each group of balanced rescue planning paths / extreme rescue planning paths, and the distance from the accident rescue points to the safety points are counted in turn, and the ratio of the number of accident rescue points included in the balanced rescue planning paths / extreme rescue planning paths to the number of accident rescue points included in the balanced rescue simulation scenarios / extreme rescue planning paths is used as the rescue safety degree, and the ratio of the path distance from the accident rescue points to the safety points included in the balanced rescue planning paths / extreme rescue planning paths to the distance of the balanced rescue planning paths / extreme rescue planning paths is used as the rescue efficiency degree;

[0032] Performing stability analysis on the balanced rescue simulation scenario / extreme rescue simulation scenario to obtain the scenario stability of the balanced rescue simulation scenario / extreme rescue simulation scenario;

[0033] Based on the scene stability, index weights are set for the rescue safety and rescue efficiency. The index weight calculation formula for the rescue safety is:

[0034]

[0035] The calculation formula of the index weight of the rescue efficiency is:

[0036]

[0037] In the formula, α represents the index weight of rescue safety, β represents the index weight of rescue efficiency, P represents the scene stability, and b represents the constant coefficient;

[0038] The rescue safety and rescue efficiency of each group of balanced rescue planning paths / extreme rescue planning paths in the balanced rescue simulation scenario / extreme rescue simulation scenario are weighted and summed using the index weights to obtain the path rescue evaluation index of each group of balanced rescue planning paths / extreme rescue planning paths. The calculation formula of the path evaluation index is:

[0039] S = α*Y + β*Z;

[0040] In the formula, S represents the path evaluation index, Y represents the rescue safety, and Z represents the rescue efficiency;

[0041] The path rescue evaluation index of each group of balanced rescue planning paths / extreme rescue planning paths is used as the simulation test evaluation value of each group of path planning algorithms in the balanced rescue simulation scenario / extreme rescue simulation scenario.

[0042] As a preferred solution of the present invention, the method of adaptively matching multiple groups of rescue planning algorithms to rescue simulation scenarios based on simulation test evaluation values ​​includes:

[0043] The path planning algorithm with the maximum simulation test evaluation value in the balanced rescue simulation scenario / extreme rescue simulation scenario is used as the balanced rescue planning algorithm / extreme rescue planning algorithm.

[0044] As a preferred solution of the present invention, before the principal component analysis, the geological structure characteristics, mine structure characteristics, accident cause characteristics, accident site personnel distribution characteristics, mine task progress characteristics, rescue channel structure characteristics, accident site meteorological characteristics and accident site structure characteristics are subjected to feature data normalization processing.

[0045] As a preferred solution of the present invention, the present invention provides a simulation test system according to the mine rescue robot path planning simulation test method, comprising:

[0046] A feature processing module is used to obtain multiple groups of historical mine accident scenes based on the big data of historical mine accidents, extract scene features of the historical mine accident scenes, and then perform discrete analysis on multiple groups of historical mine accidents based on the scene features to obtain rescue scene features;

[0047] A scene processing module is used to construct a rescue simulation scene according to the rescue scene characteristics, and mark the accident rescue points and safety points of the mine rescue robot in the rescue simulation scene;

[0048] A path evaluation module is used to perform rescue path planning based on multiple groups of path planning algorithms of the mine rescue robot based on rescue simulation scenarios to obtain multiple groups of rescue planning paths, and to perform safety and efficiency analysis on the multiple groups of rescue planning paths to obtain simulation test evaluation values ​​of each group of path planning algorithms, so as to implement scenario simulation testing of the path planning function of the mine rescue robot;

[0049] The scene matching module is used to adaptively match multiple groups of rescue planning algorithms to rescue simulation scenes based on simulation test evaluation values, so as to realize multi-scenario rescue operations of mine rescue robots.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] The present invention utilizes the extraction of scene features of historical mine accident scenes, and then performs discrete analysis on multiple groups of historical mine accidents based on the scene features to obtain rescue scene features, thereby avoiding the subjectivity and randomness of artificial settings, and at the same time obtains balanced rescue scene features and extreme rescue scene features from the rescue scene features to construct balanced rescue simulation scenes and extreme rescue simulation scenes, thereby achieving accurate satisfaction of simulation test needs with a limited number of rescue simulation scenes, avoiding invalid simulation, and performing adaptive matching of rescue planning algorithms in rescue simulation scenes, thereby realizing multi-scenario rescue operations of mine rescue robots and improving the scene rescue accuracy of mine robots. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0053] Figure 1 A flow chart of a mine rescue robot path planning simulation test method provided by an embodiment of the present invention;

[0054] Figure 2 This is a structural block diagram of a simulation test system provided by an embodiment of the present invention.

[0055] The numbers in the figure represent the following:

[0056] 1-Feature processing module; 2-Scene processing module; 3-Path evaluation module; 4-Scene matching module. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] like Figure 1 As shown, the present invention provides a mine rescue robot path planning simulation test method, comprising the following steps:

[0059] Step S1, obtaining multiple groups of historical mine accident scenes based on the big data of historical mine accidents, extracting scene features of the historical mine accident scenes, and then performing discrete analysis on the multiple groups of historical mine accidents based on the scene features to obtain rescue scene features;

[0060] Extract scene features of historical mine accident scenes, including:

[0061] In each historical accident scene of the mine, the geological structure characteristics, mine structure characteristics, accident cause characteristics, accident site personnel distribution characteristics, mine task progress characteristics, rescue channel structure characteristics, accident site meteorological characteristics, and accident site structure characteristics are extracted in turn;

[0062] The principal component analysis is performed on the geological structure characteristics, the mine structure characteristics, the accident cause characteristics, the accident site personnel distribution characteristics, the mine task progress characteristics, the rescue channel structure characteristics, the accident site meteorological characteristics and the accident site structure characteristics to obtain the information contribution value of each feature item in turn. The principal component analysis is a conventional feature analysis technology, which will not be described in detail in this embodiment.

[0063] All feature items whose information contribution values ​​are higher than the preset threshold are regarded as scene features.

[0064] Before principal component analysis, the characteristic data of geological structure characteristics, mine structure characteristics, accident cause characteristics, accident site personnel distribution characteristics, mine task progress characteristics, rescue channel structure characteristics, accident site meteorological characteristics and accident site structure characteristics were normalized.

[0065] Discrete analysis of multiple groups of historical mine accidents based on scene characteristics is performed to obtain rescue scene characteristics, including:

[0066] The expected value of the rescue scene characteristics is set, and the expected value of the rescue scene characteristics and the mine historical accident scene characteristics of multiple groups of mine historical accidents are discretely quantified to obtain the scene feature discreteness. The calculation formula of the scene feature discreteness is:

[0067]

[0068] In the formula, σ X Characterized as scene feature discreteness, X i The vector form of the mine accident scene characteristics representing the i-th mine accident, X E The vector form of the expected value of the rescue scene features, |X i -X E | Characterized by X i and X E The Euclidean distance of , N represents the total number of historical accidents in the mine;

[0069] Construct the solution constraint conditions of the expected value of the rescue scene characteristics. The solution constraint conditions are:

[0070] X E,j ∈[min(X (1,N),j ),max(X (1,N),j )];

[0071] In the formula, X E,j Characterized by X E The jth eigencomponent in X (1,N) The mine accident scene characteristics represented by the 1st to Nth mine accident history are represented by X (1,N) The jth characteristic component in is represented by min(X (1,N),j )、max(X (1,N),j ) are represented as (1,N),j A function for obtaining the minimum value of the j-th characteristic component and a function for obtaining the maximum value of the j-th characteristic component;

[0072] Based on solving the constraints, the discreteness of the scene features is minimized to obtain the equilibrium expected value of the rescue scene features, and based on solving the constraints, the discreteness of the scene features is maximized to obtain the extreme expected value of the rescue scene features. Among them, the equilibrium expected value of the rescue scene features is a balanced expression of the mine historical accident scene features of multiple mine historical accidents, and the extreme expected value of the rescue scene features is an extreme expression of the mine historical accident scene features of multiple mine historical accidents.

[0073] If the rescue simulation scenario is more similar to more of the selected mine historical accident scenarios, the discreteness between the rescue scenario characteristics of the rescue simulation scenario and the mine historical accident scenario characteristics of all the selected mine historical accident scenarios is lower, further indicating that the rescue simulation scenario is a simulation expression of the scenario with a higher frequency of occurrence of mine historical accident scenarios, which is a balanced expression; and if the rescue simulation scenario is more similar to fewer of the selected mine historical accident scenarios, the discreteness between the rescue scenario characteristics of the rescue simulation scenario and the mine historical accident scenario characteristics of all the selected mine historical accident scenarios is higher, further indicating that the rescue simulation scenario is a simulation expression of the scenario with a lower frequency of occurrence of mine historical accident scenarios, which is an extreme expression.

[0074] When simulating rescue scenarios, it is necessary to simulate conventional scenarios, that is, to simulate high-incidence historical mine accident scenarios, to test the conventional rescue effects of rescue robots in conventional accident scenarios, and conventional accident scenarios are balanced expressions of historical mine accident scenarios. It is also necessary to simulate extreme scenarios to test the extreme rescue effects of rescue robots in extreme accident scenarios, and extreme accident scenarios are extreme expressions of historical mine accident scenarios. By obtaining the conventional rescue effects and extreme rescue effects of the rescue robot, the rescue effects of the rescue robot can be fully mastered, and there is no need to change the characteristics of each scene one by one to obtain countless rescue simulation scenarios and conduct disordered and redundant analysis. Therefore, this embodiment discretely analyzes to obtain balanced rescue simulation scenarios and extreme rescue simulation scenarios, and obtains rescue simulation scenarios with a limited number of directional and accurate methods to comprehensively test the rescue effects of the rescue robot, thereby improving test efficiency.

[0075] Step S2: construct a rescue simulation scene according to the rescue scene characteristics, and mark the accident rescue points and safety points of the mine rescue robot in the rescue simulation scene;

[0076] A rescue simulation scenario is constructed based on the characteristics of the rescue scenario, including:

[0077] A balanced rescue simulation scenario is constructed based on the balanced expected value of the rescue scenario characteristics, and an extreme rescue simulation scenario is constructed based on the extreme expected value of the rescue scenario characteristics.

[0078] The accident rescue points and safety points of the mine rescue robot are marked in the rescue simulation scene, including:

[0079] In the balanced rescue simulation scenario and the extreme rescue simulation scenario, the distribution points of the personnel at the accident scene are marked as accident rescue points, and in the balanced rescue simulation scenario and the extreme rescue simulation scenario, the exits of the rescue passages are marked as safety points.

[0080] Step S3, performing rescue path planning on multiple groups of path planning algorithms of the mine rescue robot based on the rescue simulation scenario to obtain multiple groups of rescue planning paths, performing safety and efficiency analysis on the multiple groups of rescue planning paths to obtain simulation test evaluation values ​​of each group of path planning algorithms, so as to implement scenario simulation testing on the path planning function of the mine rescue robot;

[0081] The multiple path planning algorithms of the mine rescue robot are used to plan rescue paths based on the rescue simulation scenario to obtain multiple rescue planning paths. The multiple path planning algorithms are customized by the user, including:

[0082] Based on the balanced rescue simulation scenario, multiple groups of path planning algorithms of the mine rescue robot are sequentially used to plan rescue paths to obtain multiple groups of balanced rescue planning paths;

[0083] The multiple path planning algorithms of the mine rescue robot are used to plan rescue paths in sequence based on extreme rescue simulation scenarios to obtain multiple groups of extreme rescue planning paths.

[0084] The safety and efficiency of multiple rescue planning paths are analyzed to obtain simulation test evaluation values ​​of each path planning algorithm, including:

[0085] The number of accident rescue points and the distance from the accident rescue points to the safe points in each group of balanced rescue planning paths are counted in turn. The ratio of the number of accident rescue points in the balanced rescue planning path to the number of accident rescue points in the balanced rescue simulation scenario is used as the rescue safety degree. The ratio of the path distance from the accident rescue point to the safe point in the balanced rescue planning path to the balanced rescue planning path distance is used as the rescue efficiency degree.

[0086] The number of accident rescue points and the distance from the accident rescue points to the safety points included in each group of extreme rescue planning paths are counted in turn. The ratio of the number of accident rescue points included in the extreme rescue planning paths to the number of accident rescue points included in the extreme rescue simulation scenarios is taken as the rescue safety degree, and the ratio of the path distance from the accident rescue points to the safety points included in the extreme rescue planning paths to the extreme rescue planning path distance is taken as the rescue efficiency degree.

[0087] The higher the rescue safety, the more accident rescue points the rescue planning path can reach, that is, the more trapped people can be rescued. The higher the rescue efficiency, the shorter the invalid distance of the extreme rescue planning path, that is, the faster all trapped people can be rescued.

[0088] The balanced rescue simulation scenario / extreme rescue simulation scenario is subjected to stability analysis to obtain the scenario stability of the balanced rescue simulation scenario / extreme rescue simulation scenario. The stability analysis is conducted on indicators such as the structural stiffness and structural degrees of freedom of the rescue simulation scenario.

[0089] Based on the scene stability, the index weights are set for rescue safety and rescue efficiency. The calculation formula for the index weight of rescue safety is:

[0090]

[0091] The calculation formula of the index weight of rescue efficiency is:

[0092]

[0093] In the formula, α represents the index weight of rescue safety, β represents the index weight of rescue efficiency, P represents the scene stability, and b represents the constant coefficient;

[0094] The rescue safety and rescue efficiency of each group of balanced rescue planning paths / extreme rescue planning paths in the balanced rescue simulation scenario / extreme rescue simulation scenario are weighted and summed using the indicator weights to obtain the path rescue evaluation index of each group of balanced rescue planning paths / extreme rescue planning paths. The calculation formula of the path evaluation index is:

[0095] S = α*Y + β*Z;

[0096] In the formula, S represents the path evaluation index, Y represents the rescue safety, and Z represents the rescue efficiency;

[0097] The path rescue evaluation index of each group of balanced rescue planning paths / extreme rescue planning paths is used as the simulation test evaluation value of each group of path planning algorithms in the balanced rescue simulation scenario / extreme rescue simulation scenario.

[0098] When the scene stability of the balanced rescue simulation scenario / extreme rescue simulation scenario is higher, at this time, the structure of the balanced rescue simulation scenario / extreme rescue simulation scenario is more stable, and the rescue safety needs to be considered more in the rescue safety and rescue efficiency, that is, rescuing more trapped persons; when the scene stability of the balanced rescue simulation scenario / extreme rescue simulation scenario is lower, at this time, the structure of the balanced rescue simulation scenario / extreme rescue simulation scenario is more unstable, and the rescue efficiency needs to be considered more in the rescue safety and rescue efficiency, that is, rescuing trapped persons as quickly as possible.

[0099] Therefore, this embodiment establishes the index weights of rescue safety and rescue efficiency on the basis of scene stability. When the scene stability of the balanced rescue simulation scenario / extreme rescue simulation scenario is higher, the rescue safety is more important, and a higher weight is assigned to the rescue safety, while a lower weight is assigned to the rescue efficiency. When the scene stability of the balanced rescue simulation scenario / extreme rescue simulation scenario is lower, the rescue efficiency is more important, and a higher weight is assigned to the rescue efficiency, while a lower weight is assigned to the rescue safety. Moreover, the rescue efficiency and the rescue safety are in a dynamic balance with each other increasing and decreasing when the scene stability changes, presenting a trade-off between the two indicators.

[0100] Step S4: Adaptively match multiple groups of rescue planning algorithms to rescue simulation scenarios based on simulation test evaluation values ​​to achieve multi-scenario rescue operations of mine rescue robots.

[0101] Based on the simulation test evaluation values, multiple groups of rescue planning algorithms are adaptively matched to the rescue simulation scenarios, including:

[0102] The path planning algorithm with the maximum simulation test evaluation value in the balanced rescue simulation scenario / extreme rescue simulation scenario is used as the balanced rescue planning algorithm / extreme rescue planning algorithm.

[0103] like Figure 2 As shown, based on the mine rescue robot path planning simulation test method, the present invention provides a simulation test system, including:

[0104] Feature processing module 1, used to obtain multiple groups of historical mine accident scenes based on the big data of historical mine accidents, extract scene features of the historical mine accident scenes, and then perform discrete analysis on the multiple groups of historical mine accidents based on the scene features to obtain rescue scene features;

[0105] The scene processing module 2 is used to construct a rescue simulation scene according to the rescue scene characteristics, and mark the accident rescue points and safety points of the mine rescue robot in the rescue simulation scene;

[0106] The path evaluation module 3 is used to perform rescue path planning on multiple groups of path planning algorithms of the mine rescue robot based on the rescue simulation scenario to obtain multiple groups of rescue planning paths, and to perform safety and efficiency analysis on the multiple groups of rescue planning paths to obtain simulation test evaluation values ​​of each group of path planning algorithms, so as to realize scenario simulation testing on the path planning function of the mine rescue robot;

[0107] The scene matching module 4 is used to adaptively match multiple groups of rescue planning algorithms to rescue simulation scenes based on simulation test evaluation values, so as to realize multi-scenario rescue operations of mine rescue robots.

[0108] The present invention utilizes the extraction of scene features of historical mine accident scenes, and then performs discrete analysis on multiple groups of historical mine accidents based on the scene features to obtain rescue scene features, thereby avoiding the subjectivity and randomness of artificial settings, and at the same time obtains balanced rescue scene features and extreme rescue scene features from the rescue scene features to construct balanced rescue simulation scenes and extreme rescue simulation scenes, thereby achieving accurate satisfaction of simulation test needs with a limited number of rescue simulation scenes, avoiding invalid simulation, and performing adaptive matching of rescue planning algorithms in rescue simulation scenes, thereby realizing multi-scenario rescue operations of mine rescue robots and improving the scene rescue accuracy of mine robots.

[0109] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A mine rescue robot path planning simulation test method, characterized in that: The following steps are involved: Step S1, obtaining multiple groups of historical mine accident scenes based on the big data of historical mine accidents, extracting scene features of the historical mine accident scenes, and then performing discrete analysis on the multiple groups of historical mine accidents based on the scene features to obtain rescue scene features; Step S2: construct a rescue simulation scene according to the rescue scene characteristics, and mark the accident rescue points and safety points of the mine rescue robot in the rescue simulation scene; Step S3, performing rescue path planning on multiple groups of path planning algorithms of the mine rescue robot based on the rescue simulation scenario to obtain multiple groups of rescue planning paths, performing safety and efficiency analysis on the multiple groups of rescue planning paths to obtain simulation test evaluation values ​​of each group of path planning algorithms, so as to implement scenario simulation testing on the path planning function of the mine rescue robot; Step S4: adaptively matching multiple groups of rescue planning algorithms to rescue simulation scenarios based on simulation test evaluation values ​​to achieve multi-scenario rescue operations of the mine rescue robot; The extraction of scene features of historical mine accident scenes includes: In each historical accident scene of the mine, the geological structure characteristics, mine structure characteristics, accident cause characteristics, accident site personnel distribution characteristics, mine task progress characteristics, rescue channel structure characteristics, accident site meteorological characteristics, and accident site structure characteristics are extracted in turn; Principal component analysis is performed on the geological structure characteristics, mine structure characteristics, accident cause characteristics, accident site personnel distribution characteristics, mine task progress characteristics, rescue channel structure characteristics, accident site meteorological characteristics and accident site structure characteristics to obtain the information contribution value of each feature item in turn; All feature items with information contribution values ​​higher than a preset threshold are regarded as scene features; The discrete analysis of multiple groups of historical mine accidents based on scene characteristics to obtain rescue scene characteristics includes: The expected value of the rescue scene feature is set, and the expected value of the rescue scene feature and the mine historical accident scene features of multiple groups of mine historical accidents are discretely quantified to obtain the scene feature discreteness. The calculation formula of the scene feature discreteness is: In the formula, σ X Characterized as scene feature discreteness, X i The vector form of the mine accident scene characteristics representing the i-th mine accident, X E The vector form of the expected value of the rescue scene features, |X i -X E | Characterized by X i and X E The Euclidean distance of , N represents the total number of historical accidents in the mine; Constructing the solution constraint condition of the expected value of the rescue scene feature, the solution constraint condition is: X E,j ∈[min(X (1,N),j ),max(X (1,N),j )]; In the formula, X E,j Characterized by X E The jth eigencomponent in X (1,N) The mine accident scene characteristics represented by the 1st to Nth mine accident history are represented by X (1,N) The jth characteristic component in is represented by min(X (1,N),j )、max(X (1,N),j ) are represented as (1,N),j A function for obtaining the minimum value of the j-th characteristic component and a function for obtaining the maximum value of the j-th characteristic component; Based on the solving constraint condition, the discreteness of the scene feature is minimized to obtain the equilibrium expected value of the rescue scene feature, and based on the solving constraint condition, the discreteness of the scene feature is maximized to obtain the extreme expected value of the rescue scene feature, wherein the equilibrium expected value of the rescue scene feature is a balanced expression of the scene features of the mine historical accidents of multiple mine historical accidents, and the extreme expected value of the rescue scene feature is an extreme expression of the scene features of the mine historical accidents of multiple mine historical accidents; The rescue simulation scenario is constructed according to the rescue scenario characteristics, including: A balanced rescue simulation scenario is constructed based on the balanced expected value of the rescue scenario characteristics, and an extreme rescue simulation scenario is constructed based on the extreme expected value of the rescue scenario characteristics.

2. A mine rescue robot path planning simulation test method according to claim 1, characterized in that: The rescue simulation scene marks the accident rescue points and safety points of the mine rescue robot, including: In the balanced rescue simulation scenario and the extreme rescue simulation scenario, the distribution points of the personnel at the accident scene are marked as accident rescue points, and in the balanced rescue simulation scenario and the extreme rescue simulation scenario, the exits of the rescue passages are marked as safety points.

3. A mine rescue robot path planning simulation test method according to claim 2, characterized in that: The method of performing rescue path planning based on the rescue simulation scenario using the multiple path planning algorithms of the mine rescue robot to obtain multiple rescue planning paths includes: The multiple path planning algorithms of the mine rescue robot are used to plan rescue paths in sequence based on the balanced rescue simulation scenario to obtain multiple sets of balanced rescue planning paths; The multiple path planning algorithms of the mine rescue robot are used to plan rescue paths in sequence based on extreme rescue simulation scenarios to obtain multiple groups of extreme rescue planning paths.

4. A mine rescue robot path planning simulation test method according to claim 3, characterized in that: The safety and efficiency analysis of multiple groups of rescue planning paths is performed to obtain simulation test evaluation values ​​of each group of path planning algorithms, including: The number of accident rescue points included in each group of balanced rescue planning paths / extreme rescue planning paths, and the distance from the accident rescue points to the safety points are counted in turn, and the ratio of the number of accident rescue points included in the balanced rescue planning paths / extreme rescue planning paths to the number of accident rescue points included in the balanced rescue simulation scenarios / extreme rescue planning paths is used as the rescue safety degree, and the ratio of the path distance from the accident rescue points to the safety points included in the balanced rescue planning paths / extreme rescue planning paths to the distance of the balanced rescue planning paths / extreme rescue planning paths is used as the rescue efficiency degree; Performing stability analysis on the balanced rescue simulation scenario / extreme rescue simulation scenario to obtain the scenario stability of the balanced rescue simulation scenario / extreme rescue simulation scenario; Based on the scene stability, index weights are set for the rescue safety and rescue efficiency. The index weight calculation formula for the rescue safety is: The calculation formula of the index weight of the rescue efficiency is: In the formula, α represents the index weight of rescue safety, β represents the index weight of rescue efficiency, P represents the scene stability, and b represents the constant coefficient; The rescue safety and rescue efficiency of each group of balanced rescue planning paths / extreme rescue planning paths in the balanced rescue simulation scenario / extreme rescue simulation scenario are weighted and summed using the index weights to obtain the path rescue evaluation index of each group of balanced rescue planning paths / extreme rescue planning paths. The calculation formula of the path rescue evaluation index is: S = α*Y + β*Z; In the formula, S represents the path rescue evaluation index, Y represents the rescue safety, and Z represents the rescue efficiency; The path rescue evaluation index of each group of balanced rescue planning paths / extreme rescue planning paths is used as the simulation test evaluation value of each group of path planning algorithms in the balanced rescue simulation scenario / extreme rescue simulation scenario.

5. A mine rescue robot path planning simulation test method according to claim 4, characterized in that: The method of adaptively matching multiple groups of rescue planning algorithms to rescue simulation scenarios based on simulation test evaluation values ​​includes: The path planning algorithm with the maximum simulation test evaluation value in the balanced rescue simulation scenario / extreme rescue simulation scenario is used as the balanced rescue planning algorithm / extreme rescue planning algorithm.

6. A mine rescue robot path planning simulation test method according to claim 5, characterized in that: Before the principal component analysis, the geological structure characteristics, mine structure characteristics, accident cause characteristics, accident site personnel distribution characteristics, mine task progress characteristics, rescue channel structure characteristics, accident site meteorological characteristics and accident site structure characteristics are subjected to feature data normalization processing.

7. A simulation test system for the mine rescue robot path planning simulation test method according to any one of claims 1 to 6, characterized in that: include: A feature processing module (1) is used to obtain multiple groups of historical mine accident scenes based on the big data of historical mine accidents, extract scene features of the historical mine accident scenes, and then perform discrete analysis on the multiple groups of historical mine accidents based on the scene features to obtain rescue scene features; A scene processing module (2) is used to construct a rescue simulation scene according to the rescue scene characteristics, and mark the accident rescue point and safety point of the mine rescue robot in the rescue simulation scene; A path evaluation module (3) is used to perform rescue path planning on multiple groups of path planning algorithms of the mine rescue robot based on a rescue simulation scenario to obtain multiple groups of rescue planning paths, and to perform safety and efficiency analysis on the multiple groups of rescue planning paths to obtain a simulation test evaluation value of each group of path planning algorithms, so as to implement a scenario-based simulation test on the path planning function of the mine rescue robot; The scene matching module (4) is used to adaptively match multiple groups of rescue planning algorithms to rescue simulation scenes based on simulation test evaluation values, so as to realize multi-scenario rescue operations of the mine rescue robot.

Images