A regional defense method based on defense knowledge model library

Abstract

The present invention relates to a regional defense method based on a defense knowledge model library, and belongs to the field of robot technology. In the present invention, after the robot is started, the robot control software connects to the robot and starts the map service; the operator draws the area on the map according to the business needs, and the robot control software generates the duty plan and the patrol plan according to the drawn area, and binds the duty plan and the patrol plan into a defense strategy according to the defense level and defense parameters; the filling color of the area is automatically generated according to different defense levels and the defense strategy is called; the Hungarian algorithm is called to assign a specified number of duty robots to work at fixed points, and a specified number of patrol robots to patrol on the regional road network. The robot performs the task, and the robot control software optimizes the defense knowledge model library. Through the scheme of the present invention, the operator performs regional defense according to the prompts of the defense knowledge model library, which greatly improves the work efficiency and enhances the intelligent level of defense management.

Description

Technical Field

[0001] The invention belongs to the field of robot technology, and in particular relates to a regional defense method based on a defense knowledge model library. Background Art

[0002] With the development of society and the advancement of science and technology, security and defense issues have become increasingly prominent. Traditional regional defense methods can no longer meet the current complex and changing security environment. They rely on manpower for patrolling and monitoring, which is inefficient and easily affected by human factors. First, when manually deploying robot tasks, the actual scene is complex and task allocation is difficult; second, manual planning has a high time cost and cannot meet the needs of rapid response to emergencies. Third, there are many defense parameter settings, and optimization and upgrading are difficult. Summary of the invention

[0003] 1. Technical issues to be resolved

[0004] The technical problem to be solved by the present invention is how to provide a regional defense method based on a defense knowledge model library to solve the problem that traditional regional defense means are difficult to meet the current complex and changeable security environment.

[0005] (II) Technical solution

[0006] In order to solve the above technical problems, the present invention proposes a regional defense method based on a defense knowledge model library, which includes the following steps:

[0007] S1. After the robot is started, the robot control software connects to the robot and starts the map service;

[0008] S2. The operator draws an area on the map according to business needs. The robot control software first identifies key buildings in the area, extracts fixed defense points in the area, and generates a duty plan based on the fixed defense points and defense parameters; then extracts the road network data in the area, and calls the decision tree algorithm to perform task allocation based on the binary graph of the corresponding road network data to divide the road network in the area into small areas, and then calls the intelligent path planning service to generate path planning for the small areas, and finally forms a patrol plan for the entire defense area; finally, the duty plan and patrol plan are bound into defense strategies according to the defense level and defense parameters, and stored in the defense knowledge model library of the area;

[0009] S3. The operator sets the defense level for the area, automatically generates the filling color of the area according to different defense levels, and calls the defense strategy in the defense knowledge model library;

[0010] S4. The operator calls the Hungarian algorithm to assign a specified number of duty robots to work at fixed points and a specified number of patrol robots to patrol on the regional road network according to the defense strategy in the defense knowledge model library;

[0011] S5. Robots assigned to duty tasks work at fixed locations and turn on face recognition. Robots assigned to patrol tasks turn on abnormal behavior monitoring and patrol the regional road network to detect whether there is any abnormal behavior.

[0012] S6. The robot control software determines whether to update the defense parameters based on the evaluation algorithm. If an update is required, the defense parameters and defense strategies are updated to optimize the defense knowledge model library.

[0013] (III) Beneficial effects

[0014] The present invention proposes a regional defense method based on a defense knowledge model library. The regional defense method based on a defense knowledge model library of the present invention performs regional defense in an automated and intelligent manner, thereby improving work efficiency and the intelligent level of defense management. At the same time, the application of technologies such as face recognition and abnormal behavior detection enhances the security and reliability of defense.

[0015] The present invention proposes a regional defense method based on a defense knowledge model library, and the present invention has the following beneficial effects:

[0016] For complex practical scenarios, the automatic task allocation method can generate the optimal regional defense strategy under different constraints such as time and space restrictions and the capabilities and resources of the robots.

[0017] When the regional defense level is updated, the troop deployment can directly call the defense strategy in the knowledge model library and respond quickly.

[0018] Defense parameters can be automatically generated through evaluation algorithms without manual settings, and defense strategies can be optimized and upgraded through iterations. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of the regional defense method based on the defense knowledge model library of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, content and advantages of the present invention more clear, the specific implementation methods of the present invention are further described in detail below in conjunction with the drawings and examples.

[0021] The present invention relates to a regional defense method based on a defense knowledge model library, comprising: (1) after the robot is started, the robot control software connects to the robot and starts the map service. (2) the operator draws the area on the map according to the business needs, the robot control software first identifies the key buildings in the area, extracts the fixed points of defense in the area, and generates a duty plan according to the fixed points and defense parameters. Then the road network data in the area is extracted, and according to the binary graph of the corresponding road network data, the decision tree algorithm is called to perform task allocation to realize the equal division of the road network in the area into small areas, and then the intelligent path planning service is called to generate the path planning for the small area, and finally a patrol plan for the entire defense area is formed. Finally, the duty plan and the patrol plan are bound into a defense strategy according to the defense level and defense parameters, and stored in the defense knowledge model library of the area. (3) the operator sets the defense level of the area, automatically generates the filling color of the area according to different defense levels, and calls the defense strategy in the defense knowledge model library. (4) The operator uses the Hungarian algorithm to assign a specified number of duty robots to work at fixed points and a specified number of patrol robots to patrol the regional road network based on the defense strategy in the defense knowledge model library. (5) Robots assigned to duty tasks work at fixed points and enable face recognition. (6) Robots assigned to patrol tasks enable abnormal behavior monitoring and patrol the regional road network to detect whether there is abnormal behavior. (7) The robot control software determines whether to update the defense parameters based on the evaluation algorithm. If an update is required, the defense parameters and defense strategy are updated to optimize the defense knowledge model library. The operator performs regional defense according to the prompts of the defense knowledge model library, which greatly improves work efficiency and enhances the intelligent level of defense management.

[0022] In order to solve the above technical problems, the present invention proposes a regional defense method based on a defense knowledge model library. The present invention relates to a regional defense method based on a defense knowledge model library, comprising the following steps:

[0023] S1. After the robot is started, the robot control software connects to the robot and starts the map service;

[0024] S2. The operator draws an area on the map according to business needs. The robot control software first identifies key buildings in the area, extracts fixed defense points in the area, and generates a duty plan based on the fixed defense points and defense parameters; then extracts the road network data in the area, and calls the decision tree algorithm to perform task allocation based on the binary graph of the corresponding road network data to divide the road network in the area into small areas, and then calls the intelligent path planning service to generate path planning for the small areas, and finally forms a patrol plan for the entire defense area; finally, the duty plan and patrol plan are bound into defense strategies according to the defense level and defense parameters, and stored in the defense knowledge model library of the area;

[0025] S3. The operator sets the defense level for the area, automatically generates the filling color of the area according to different defense levels, and calls the defense strategy in the defense knowledge model library.

[0026] S4. The operator calls the Hungarian algorithm to assign a specified number of duty robots to work at fixed points and a specified number of patrol robots to patrol the regional road network according to the defense strategy in the defense knowledge model library.

[0027] S5. Robots assigned to duty tasks work at fixed points and enable face recognition. Robots assigned to patrol tasks enable abnormal behavior monitoring and patrol the regional road network to detect whether there is any abnormal behavior.

[0028] S6. The robot control software determines whether to update the defense parameters based on the evaluation algorithm. If an update is required, the defense parameters and defense strategies are updated to optimize the defense knowledge model library.

[0029] Furthermore, the map service realizes the functions of basic map operations, intelligent path planning and slope analysis.

[0030] Furthermore, basic map operations include zooming in, zooming out, roaming, measuring, clearing the screen, and marking points, lines, and areas on the map.

[0031] Furthermore, intelligent path planning refers to marking the starting point, waypoints, avoidance points, avoidance areas, and target points on the map. The map service can automatically plan the robot's movement route and display it on the map.

[0032] Furthermore, slope analysis refers to analyzing the slope of a specified area on a map. When the slope is less than 15 degrees, the patrol robot can be planned to move in the area.

[0033] Furthermore, in the defense knowledge model library, there are different defense strategies according to the defense level of the area. The defense strategies include basic elements such as the number of robots required, the size of the area, the duty points, the number of duty robots, the patrol method, the number of patrol robots, the patrol route, the patrol interval, and the patrol speed.

[0034] When the defense level is 0, each duty point has 4 duty robots, and every 2 robots form a minimum duty unit to take turns to guard and charge. There are 8 patrol robots, and every 4 robots form a team. Two teams of robots form the smallest patrol unit, and they take turns to patrol and charge in parallel on the regional road network.

[0035] When the defense level is 1, each duty point has 2 duty robots, and every 2 robots form a minimum duty unit to take turns to guard and charge. There are 4 patrol robots, and every 2 robots form a team. Two teams of robots form the minimum patrol unit, and they take turns to patrol and charge in parallel on the regional road network.

[0036] When the defense level is 2, each duty point has 2 duty robots, and each 2 robots form a minimum duty unit to take turns to guard and charge. 2 patrol robots, 2 robots form a minimum patrol unit, and take turns to patrol and charge in parallel on the regional road network.

[0037] When the defense level is level 3, there is no need for duty robots. There are 2 patrol robots, and the two robots take turns patrolling and charging on the regional road network.

[0038] Furthermore, patrol modes include: follow-up patrol, parallel patrol, and building-surrounding patrol. Follow-up patrol means that multiple robots form a formation to patrol. Parallel patrol means that multiple robots patrol the road network lines in an area that is evenly divided. Building-surrounding patrol means that robots patrol the buildings in an area that is evenly divided. The default mode is parallel patrol.

[0039] Furthermore, each defense strategy in the defense knowledge model library includes two plans: patrol and duty. The generation of the patrol plan first identifies key buildings and extracts the road network data in the area. Then, based on the binary graph of the corresponding road network data, the decision tree algorithm is called to perform task allocation to divide the road network in the area into small areas. Then, the intelligent path planning service is called to generate path planning for the small areas, and finally a patrol plan for the entire defense area is formed. The generation of the duty plan first identifies key buildings in the area, extracts the fixed defense points in the area, and generates a duty plan based on the fixed defense points and defense parameters.

[0040] Furthermore, the defense strategies in the defense knowledge model library can be flexibly adjusted, such as the minimum number of robots constituting a duty unit can be set, the minimum number of patrol robots can be set, and the patrol method can be set.

[0041] Furthermore, the shape of the area can be circular, rectangular, or polygonal. The area's defense level is divided into four levels: 0, 1, 2, and 3, with 0 being the highest level. If the area's defense level is 0, the area's fill color is red; if the area's defense level is 1, the area's fill color is orange; if the area's defense level is 2, the area's fill color is yellow; and if the area's defense level is 3, the area's fill color is green.

[0042] Furthermore, abnormal behaviors include: secretly photographing, scattering, arson, climbing over walls, etc. During the abnormal behavior detection process, when it is recognized that there is an angle between the joints of a person's hands, the elbow joints are close to parallel, and a mobile phone is recognized on a pedestrian, it is judged as secretly photographing. When the key points of the elbow and hip joints are greater than ninety degrees, and the key points of the hands are basically aligned with the key points of the eyes, it is judged as scattering. If a wall is identified in the designated area, and at the same time, it is identified that the hands and feet of pedestrians appear on the wall or the pedestrians are close to the wall, it is judged as climbing over the wall. When the target recognition module detects the target person and the flame, it is judged as arson.

[0043] Furthermore, the robot control software determines whether to update the defense parameters based on the evaluation algorithm. If an update is required, the defense parameters and defense strategies are updated to optimize the defense knowledge model library.

[0044] Embodiment 1:

[0045] The present invention discloses a regional defense method based on a defense knowledge model library.

[0046] (1) After the robot is started, the robot control software connects to the robot and starts the map service.

[0047] (2) The operator draws an area on the map according to business needs. The robot control software first identifies key buildings in the area, extracts fixed defense points in the area, and generates a duty plan based on the fixed points and defense parameters. Then, the road network data in the area is extracted. According to the binary graph of the corresponding road network data, the decision tree algorithm is called to perform task allocation to divide the road network in the area into small areas. Then, the intelligent path planning service is called to generate path planning for the small areas, and finally a patrol plan for the entire defense area is formed. Finally, the duty plan and patrol plan are bound into defense strategies according to the defense level and defense parameters, and stored in the defense knowledge model library of the area.

[0048] (3) The operator sets the defense level for the area, automatically generates the fill color of the area according to the different defense levels, and calls the defense strategy in the defense knowledge model library.

[0049] (4) The operator uses the Hungarian algorithm to assign a specified number of duty robots to work at fixed points and a specified number of patrol robots to patrol the regional road network based on the defense strategy in the defense knowledge model library.

[0050] (5) Robots assigned to duty tasks will work at fixed locations and enable facial recognition.

[0051] (6) Assigning patrol tasks to the robots will enable abnormal behavior monitoring and patrol the regional road network to detect whether there is any abnormal behavior.

[0052] (7) The robot control software determines whether to update the defense parameters based on the evaluation algorithm. If an update is required, the defense parameters and defense strategies are updated to optimize the defense knowledge model library.

[0053] The robot control software development environment of the present invention is shown in Table 1. The core of the software is to complete the connection with the robot, data transmission and processing, user interaction processing, map service loading, calling and integration of intelligent algorithm components, and calling and maintaining the knowledge model library. The system adopts a B / S architecture, cooperates with the WebSocket between the front end and the back end, and the Socket long connection component between the back end and the robot. The user performs relevant information viewing and control operations on the Web interface, and sends control instructions to the robot through the server, and finally completes the robot's command and control, abnormal display, face recognition display, situation display and other functions.

[0054] Table 1 Software development environment

[0055] Software tools / environment use Version Ruby Backend development language 5.31 rails Backend development infrastructure 7.0.0 VSCode Code browsing and editing 5.4.0 WebSocket Communication Protocol 5.14.2 vue Front-end development language 3.3.4 MySql database 4.11.1

[0056] Map services implement basic map operations, intelligent path planning, and slope analysis functions.

[0057] Basic map operations include zooming in, zooming out, roaming, measuring, clearing the screen, and marking points, lines, and areas on the map.

[0058] Intelligent path planning refers to marking the starting point, waypoints, avoidance points, avoidance areas, and target points on the map. The map service can automatically plan the robot's movement route and display it on the map.

[0059] Slope analysis refers to analyzing the slope of a specified area on a map. When the slope is less than 15 degrees, the patrol robot can be planned to move in the area.

[0060] The defense knowledge model library has different defense strategies according to the defense level of the area. The defense strategies include the number of robots required, the size of the area, the duty points, the number of duty robots, the patrol method, the number of patrol robots, the patrol route, the patrol interval, the patrol speed and other basic elements.

[0061] When the defense level is 0, each duty point has 4 duty robots, and every 2 robots form a minimum duty unit to take turns to guard and charge. There are 8 patrol robots, and every 4 robots form a team. Two teams of robots form the smallest patrol unit, and they take turns to patrol and charge in parallel on the regional road network.

[0062] When the defense level is 1, each duty point has 2 duty robots, and every 2 robots form a minimum duty unit to take turns to guard and charge. There are 4 patrol robots, and every 2 robots form a team. Two teams of robots form the minimum patrol unit, and they take turns to patrol and charge in parallel on the regional road network.

[0063] When the defense level is 2, each duty point has 2 duty robots, and each 2 robots form a minimum duty unit to take turns to guard and charge. 2 patrol robots, 2 robots form a minimum patrol unit, and take turns to patrol and charge in parallel on the regional road network.

[0064] When the defense level is level 3, there is no need for duty robots. There are 2 patrol robots in total. The two robots take turns patrolling and charging on the regional road network. The patrol modes are divided into follow-up patrol, parallel patrol, and building-surrounding patrol. Follow-up patrol means that multiple robots form a formation to patrol. Parallel patrol means that multiple robots divide the road network lines in the area equally to patrol. Building-surrounding patrol means that robots divide the buildings in the area equally to patrol around. The default mode is parallel patrol.

[0065] Each defense strategy in the defense knowledge model library includes two plans: patrol and duty. The generation of the patrol plan first identifies key buildings and extracts the road network data in the area. Then, based on the binary graph of the corresponding road network data, the decision tree algorithm is called to perform task allocation to divide the road network in the area into small areas. Then, the intelligent path planning service is called to generate path planning for the small areas, and finally a patrol plan for the entire defense area is formed. The generation of the duty plan first identifies key buildings in the area, extracts the fixed defense points in the area, and generates a duty plan based on the fixed defense points and defense parameters.

[0066] The defense strategies in the defense knowledge model library can be adjusted flexibly, such as the minimum number of robots in the duty unit can be set, the minimum number of patrol robots can be set, and the patrol method can be set.

[0067] The shape of the area can be circular, rectangular, or polygonal. The area's defense level is divided into four levels: 0, 1, 2, and 3, with 0 being the highest level. If the area's defense level is 0, the area's fill color is red; if the area's defense level is 1, the area's fill color is orange; if the area's defense level is 2, the area's fill color is yellow; and if the area's defense level is 3, the area's fill color is green.

[0068] Abnormal behaviors include: secretly photographing, throwing, arson, climbing over walls, etc. During abnormal behavior detection, when it is recognized that there is an angle between the joints of a person's hands, the elbow joints are close to parallel, and a mobile phone is recognized on a pedestrian, it is judged as secretly photographing. When the key points of the elbow and hip joints are greater than 90 degrees, and the key points of the hands are basically aligned with the key points of the eyes, it is judged as throwing. If a wall is recognized in the designated area, and at the same time, it is recognized that the hands and feet of pedestrians appear on the wall or the pedestrians are close to the wall, it is judged as climbing over the wall. When the target recognition module detects the target person and the flame, it is judged as arson.

[0069] The robot control software determines whether to update the defense parameters based on the evaluation algorithm. If an update is required, the defense parameters and defense strategies are updated to optimize the defense knowledge model library.

[0070] The regional defense method based on the defense knowledge model library of the present invention performs regional defense in an automated and intelligent manner, thereby improving work efficiency and the intelligent level of defense management. At the same time, the application of technologies such as face recognition and abnormal behavior detection enhances the security and reliability of defense.

[0071] The present invention proposes a regional defense method based on a defense knowledge model library, and the present invention has the following beneficial effects:

[0072] For complex practical scenarios, the automatic task allocation method can generate the optimal task allocation plan under different constraints such as time and space restrictions and the robot's capabilities and resources.

[0073] When the regional defense level is updated, the operator can directly call the defense strategy in the knowledge model library for a quick response.

[0074] Defense parameters can be automatically generated through evaluation algorithms without manual settings, and defense strategies can be optimized and upgraded through iterations.

[0075] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A regional defense method based on a defense knowledge model library, characterized in that: The method comprises the following steps: S1. After the robot is started, the robot control software connects to the robot and starts the map service; S2. The operator draws an area on the map according to business needs. The robot control software first identifies key buildings in the area, extracts fixed defense points in the area, and generates a duty plan based on the fixed defense points and defense parameters; then extracts the road network data in the area, and calls the decision tree algorithm to perform task allocation based on the binary graph of the corresponding road network data to divide the road network in the area into small areas, and then calls the intelligent path planning service to generate path planning for the small areas, and finally forms a patrol plan for the entire defense area; finally, the duty plan and patrol plan are bound into defense strategies according to the defense level and defense parameters, and stored in the defense knowledge model library of the area; S3. The operator sets the defense level for the area, automatically generates the filling color of the area according to different defense levels, and calls the defense strategy in the defense knowledge model library; S4. The operator calls the Hungarian algorithm to assign a specified number of duty robots to work at fixed points and a specified number of patrol robots to patrol on the regional road network according to the defense strategy in the defense knowledge model library; S5. Robots assigned to duty tasks work at fixed locations and turn on face recognition. Robots assigned to patrol tasks turn on abnormal behavior monitoring and patrol the regional road network to detect whether there is any abnormal behavior. S6. The robot control software determines whether to update the defense parameters according to the evaluation algorithm. If an update is required, the defense parameters and defense strategies are updated to optimize the defense knowledge model library. in, In the defense knowledge model library, there are different defense strategies according to the defense level of the area. The defense strategies include: the number of robots required, the size of the area, the duty points, the number of duty robots, the patrol method, the number of patrol robots, the patrol route, the patrol interval and the patrol speed; When the defense level is 0, each duty point has 4 duty robots, and every 2 robots form a minimum duty unit to take turns to guard and charge; 8 patrol robots, every 4 robots form a team, two teams of robots form the smallest patrol unit, and take turns to patrol and charge in parallel on the regional road network; When the defense level is 1, each duty point has 2 duty robots, and every 2 robots form a minimum duty unit to take turns to guard and charge; 4 patrol robots, every 2 robots form a team, two teams of robots form the minimum patrol unit, take turns to patrol and charge in parallel on the regional road network; When the defense level is 2, each duty point has 2 duty robots, and every 2 robots form the smallest duty unit to take turns to guard and charge; 2 patrol robots, 2 robots form the smallest patrol unit, take turns to patrol and charge in parallel on the regional road network; When the defense level is level 3, there is no need for duty robots; there are 2 patrol robots, and the two robots take turns patrolling and charging on the regional road network; Abnormal behaviors include: secretly taking photos, throwing things, arson and climbing over walls; during the abnormal behavior detection process, when it is recognized that there is an angle between the joints of a person's hands, the elbow joints are close to parallel, and a mobile phone is recognized on the pedestrian, it is judged as secretly taking photos; when the key points of the elbow and hip joints are greater than ninety degrees, and the hand key points are basically aligned with the eye key points, it is judged as throwing things; if a wall is recognized in the designated area, and at the same time it is recognized that the hands and feet of a pedestrian appear on the wall or the pedestrian is close to the wall, it is judged as climbing over the wall; when the target recognition module detects the target person and flames, it is judged as arson.

2. The regional defense method based on the defense knowledge model library as claimed in claim 1, characterized in that: Map services include: basic map operations, intelligent route planning and slope analysis.

3. The regional defense method based on the defense knowledge model library as claimed in claim 2, characterized in that: Basic map operations include zooming in, zooming out, roaming, measuring, clearing the screen, and marking points, lines, and areas on the map.

4. The regional defense method based on the defense knowledge model library as claimed in claim 2, characterized in that: Intelligent path planning refers to marking the starting point, waypoints, avoidance points, avoidance areas, and target points on the map. The map service can automatically plan the robot's movement route and display it on the map.

5. The regional defense method based on the defense knowledge model library as claimed in claim 2, characterized in that: Slope analysis refers to analyzing the slope of a specified area on a map. When the slope is less than 15 degrees, the patrol robot can be planned to move in the area.

6. The regional defense method based on the defense knowledge model library as claimed in claim 1, characterized in that: The patrol modes include: follow-up patrol, parallel patrol, and building-surrounding patrol; follow-up patrol refers to multiple robots forming a formation to patrol; parallel patrol refers to multiple robots patrolling the road network lines in an area equally divided; building-surrounding patrol refers to robots patrolling around buildings in an area equally divided; the default mode is parallel patrol.

7. The regional defense method based on the defense knowledge model library as claimed in claim 1, characterized in that: The defense strategies in the defense knowledge model library can be adjusted flexibly, among which the minimum number of robots in the duty unit can be set, the minimum number of patrol robots can be set, and the patrol method can be set.

8. The regional defense method based on the defense knowledge model library as claimed in claim 1, characterized in that: The shape of the area is a circle, rectangle or polygon; the defense level of the area is divided into four levels: 0, 1, 2 and 3, with 0 being the highest level; if the defense level of the area is 0, the fill color of the area is red; if the defense level of the area is 1, the fill color of the area is orange; if the defense level of the area is 2, the fill color of the area is yellow; if the defense level of the area is 3, the fill color of the area is green.

Images