Virtual simulation test method for fire-fighting robot path planning performance analysis

Abstract

A virtual simulation test method for fire-fighting robot path planning performance analysis belongs to a virtual environment simulation technology, and comprises a fire-fighting rescue robot performance simulation method, a fire-fighting rescue scene virtualization simulation method and a performance analysis method. The problems of robot performance simulation, fire rescue scene virtualization simulation, performance test methods and the like are solved, and a test basis is provided for later research and development and shaping of the robot.

Description

Technical Field

[0001] This invention belongs to the field of virtual environment simulation technology, specifically relating to a virtual simulation testing method for path planning performance analysis of fire-fighting robots. Background Technology

[0002] Fire scenes are complex during fire rescue operations. Some fire rescue robots possess path planning and intelligent dispatching capabilities. These robots aim to complete missions within limited space and in the shortest possible time. Utilizing algorithms such as machine learning and deep reinforcement learning, they implement command and dispatch strategies for indoor firefighting and rescue robots. However, the following issues arose during the research and testing process: First, the scene of a fire rescue is different from a normal disaster-free scene, mainly due to the interference of smoke, which has a significant impact on robots that currently rely primarily on lidar for perception. Therefore, special adaptation to the scene is required.

[0003] Secondly, in the process of robot research and development and testing, software algorithm and hardware development are generally carried out in parallel. At intermediate milestone nodes, in order to verify the reliability of the algorithm before the hardware is completed, special test and verification methods need to be designed. Test methods based on virtual simulation are necessary. Summary of the Invention

[0004] Based on the above-mentioned technical problems, this invention provides a virtual simulation testing method for path planning performance analysis of fire-fighting robots, which solves problems such as robot performance simulation, virtual simulation of fire rescue scenarios, and testing performance methods, and provides testing basis for the later research and development and finalization of robots.

[0005] To achieve the above objectives, the main technical solutions adopted by the present invention include: A virtual simulation testing method for path planning performance analysis of firefighting robots is provided. The method includes firefighting robot performance simulation, firefighting scenario virtualization simulation, and performance analysis methods. The firefighting robot performance simulation includes setting the following parameters for the firefighting robot: (1) Horizontal detection range angle of lidar; (2) Vertical detection range angle of lidar; (3) Linear detection range of lidar under smoke-free conditions; (4) Linear detection range of lidar under smoky conditions; (5) Walking speed of the fire rescue robot; (6) Turning and rotation speed of the fire rescue robot; The virtual simulation of the fire and rescue scenario includes the following steps: Set up the following steps before starting the test: Step 1: Specify the start and end points; Step 2: Randomly generate 10 circular smoke dots of 1m x 1m on the map as simulated interference sources to interfere with lidar detection; Step 3: Randomly generate 10 square obstacles of 0.5m x 0.5m on the map as simulated obstacles to interfere with the movement of the fire rescue robot; The performance analysis method includes the following steps: Step 1: Set a fixed start point and end point, randomly set interference sources and obstacles, and record the time t from robot start-up to reaching the end point; Step 2: Repeat the test 10 times and record the arrival time for each of the 10 tests as t1, t2, t3, t4, t5, t6, t7, t8, t9, t10; Step 3: Record the maximum value of the 10 time records as tmax and the minimum value as tmin; Step 4: Use the average arrival time T after removing the maximum and minimum values ​​as the performance evaluation standard, i.e., T = (t1 + t2 + t3 + t4 + t5 + t6 + t7 + t8 + t9 + t10 - tmax - tmin) / 8.

[0006] Specifically, the performance simulation of the fire rescue robot is for fire rescue robots that use lidar as a sensing method.

[0007] Specifically, the fire rescue scenario virtualization simulation sets the rescue scenario as a single-story building scene of 1,000 square meters, with an atrium and multiple rooms.

[0008] Specifically, in the performance analysis method, the average arrival time T after testing different algorithms and different robots is used as an evaluation standard for the algorithm's advancement and the device's hardware performance.

[0009] The beneficial effects of this invention are: This invention utilizes virtual environment simulation technology to design a quantitative test method for path planning performance analysis of fire-fighting robots, focusing on solving the problem of the lack of standardized test methods in the early design process of fire-fighting robots, so as to achieve the goal of designing in a way that takes into account scientificity, comfort and practicality, and to provide test basis for the later research and development and finalization of the robot. Detailed Implementation

[0010] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below through specific embodiments.

[0011] This invention provides a virtual simulation testing method for path planning performance analysis of firefighting robots. The purpose is to use virtual environment simulation technology to design a testing method for path planning performance analysis of firefighting robots, focusing on solving the problem of the lack of standardized testing methods in the early design process of firefighting robots, and providing testing basis for the later research and development and finalization of robots.

[0012] The virtual simulation testing method for path planning performance analysis of fire-fighting robots of the present invention includes three aspects: performance simulation of fire-fighting rescue robots, virtual simulation of fire-fighting rescue scenarios, and performance analysis methods.

[0013] Part One: Performance Simulation of the Firefighting and Rescue Robot. This invention primarily focuses on firefighting and rescue robots that utilize lidar as their sensing method. Therefore, the following parameters for the firefighting and rescue robot need to be set: (1) Horizontal detection range angle of lidar; (2) Vertical detection range angle of lidar; (3) Linear detection range of lidar under smoke-free conditions; (4) Linear detection range of lidar under smoky conditions; (5) Walking speed of the fire rescue robot; (6) Turning and rotation speed of the fire rescue robot; Part Two: Virtual Simulation of Fire and Rescue Scenarios. The rescue scenario is set as a 1000-square-meter single-story building with an atrium and multiple rooms. The following steps are set up before starting the test: Step 1: Specify the start and end points; Step 2: Randomly generate 10 circular smoke dots of 1m*1m on the map as simulated interference sources to interfere with lidar detection.

[0014] Step 3: Randomly generate 10 square obstacles of 0.5m x 0.5m on the map as simulated obstacles to interfere with the movement of the fire rescue robot.

[0015] Part Three, Performance Analysis Methodology, includes the following steps: Step 1: Set a fixed start point and end point, randomly set interference sources and obstacles, and record the time t from robot start-up to reaching the end point.

[0016] Step 2: Repeat the test 10 times and record the arrival time for each of the 10 tests as t1, t2, t3, t4, t5, t6, t7, t8, t9, and t10.

[0017] Step 3: Record the maximum value of the 10 time records as tmax and the minimum value as tmin.

[0018] Step 4: Use the average arrival time T after removing the maximum and minimum values ​​as the performance evaluation standard, i.e., T = (t1 + t2 + t3 + t4 + t5 + t6 + t7 + t8 + t9 + t10 - tmax - tmin) / 8.

[0019] The average arrival time T after testing different algorithms and different robots serves as an evaluation standard for the algorithm's advancement and the device's hardware performance; the shorter the time, the better the effect.

[0020] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Any modifications, alterations, substitutions, and variations made by those skilled in the art to the above embodiments are within the scope of the present invention.

Claims

1. A virtual simulation testing method for performance analysis of path planning in firefighting robots, characterized in that, The method includes performance simulation of fire rescue robots, virtual simulation of fire rescue scenarios, and performance analysis methods; the performance simulation of fire rescue robots includes setting the following parameters for the fire rescue robot: (1) Horizontal detection range angle of lidar; (2) Vertical detection range angle of lidar; (3) Linear detection range of lidar under smoke-free conditions; (4) Linear detection range of lidar under smoky conditions; (5) Walking speed of the fire rescue robot; (6) Turning and rotation speed of the fire rescue robot; The virtual simulation of the fire and rescue scenario includes the following steps: Set up the following steps before starting the test: Step 1: Specify the start and end points; Step 2: Randomly generate 10 circular smoke dots of 1m x 1m on the map as simulated interference sources to interfere with lidar detection; Step 3: Randomly generate 10 square obstacles of 0.5m x 0.5m on the map as simulated obstacles to interfere with the movement of the fire rescue robot; The performance analysis method includes the following steps: Step 1: Set a fixed start point and end point, randomly set interference sources and obstacles, and record the time t from robot start-up to reaching the end point; Step 2: Repeat the test 10 times and record the arrival time for each of the 10 tests as t1, t2, t3, t4, t5, t6, t7, t8, t9, t10; Step 3: Record the maximum value of the 10 time records as tmax and the minimum value as tmin; Step 4: Use the average arrival time T after removing the maximum and minimum values ​​as the performance evaluation standard, i.e., T = (t1 + t2 + t3 + t4 + t5 + t6 + t7 + t8 + t9 + t10 - tmax - tmin) / 8.

2. The virtual simulation testing method for path planning performance analysis of firefighting robots according to claim 1, characterized in that, The performance simulation of the fire rescue robot is for fire rescue robots that use lidar as their sensing method.

3. The virtual simulation testing method for path planning performance analysis of firefighting robots according to claim 1, characterized in that, The virtual simulation of the fire rescue scenario sets the rescue scenario as a single-story building scene of 1,000 square meters, with an atrium and multiple rooms.

4. The virtual simulation testing method for path planning performance analysis of firefighting robots according to claim 1, characterized in that, In the performance analysis method, the average arrival time T after testing different algorithms and different robots is used as the evaluation standard for algorithm advancement and equipment hardware performance.