Autonomous path planning function test system and method suitable for lunar surface detector

Abstract

The invention discloses an autonomous path planning function test system and method suitable for a lunar surface probe, and the system comprises a lunar surface scene simulation module, a dynamics module, a sensor module, a controller module, a data collection and storage module, and a test evaluation module, and the above modules work cooperatively through a closed-loop data flow. According to the test method, a lunar surface environment model, a detector dynamics module and a sensor model are deployed on a real-time computing platform, and a detector autonomous path planning system can be operated in real time in a simulation environment. The test method provided by the invention can shorten the research and development period of the autonomous path planning system of the lunar surface detector, reduce the technical risk, improve the success rate of the task, and provide technical support for smooth implementation of the future lunar surface detection task.

Description

Technical Field

[0001] This invention relates to a testing system and method for autonomous path planning functionality of a lunar probe, particularly for the verification and evaluation of such functionality. The testing system employs a distributed computing architecture to achieve efficient collaboration within the system, aiming to verify the path planning capabilities, adaptability, and accuracy of a lunar probe in the complex lunar environment. Background Technology

[0002] With the continuous development of deep space exploration technology, lunar probes have become key equipment for carrying out lunar exploration missions. Autonomous path planning is one of the core functions of a lunar probe, enabling it to autonomously determine and plan a suitable path based on environmental information, thereby achieving efficient and accurate mission completion. However, due to the complex and ever-changing lunar environment, with many uncertainties such as terrain undulations, obstacle distribution, and lighting variations, autonomous path planning poses significant challenges to the probe.

[0003] In the design and development of an autonomous path planning system for a lunar probe, the testing and verification of the path planning function are typically scheduled for the later stages of the development process. At this point, potential problems from the design phase may be difficult to identify and correct in a timely manner. If errors in the path planning algorithm or system logic are discovered during later testing, the correction process often requires significant time and resources, and may even lead to delays in the entire project, severely impacting mission execution and schedule. Furthermore, directly deploying the lunar probe for testing in actual missions may encounter problems such as complex terrain, sensor errors, or path planning failures, causing the probe to fail to follow the expected path, increasing the risk of mission failure, and potentially damaging the probe, resulting in unpredictable property losses and resource waste.

[0004] Given the above reasons, how to safely, quickly, and effectively test and verify the path planning system in the early stages of lunar probe development, and ensure the system's reliability and safety in actual missions, has become an urgent technical challenge.

[0005] To ensure the probe can effectively address these challenges during its lunar mission, its path planning capabilities must be thoroughly evaluated through testing and verification. This process requires not only simulating various changes in the lunar environment but also considering the probe's dynamic characteristics and the accuracy of sensor data during actual operation. Existing lunar probe path planning test platforms typically have shortcomings, such as a lack of dynamic simulation capabilities for complex environments, an inability to effectively integrate multi-source data for real-time evaluation, and relatively limited test scenarios. Therefore, developing a test system suitable for the autonomous path planning function of a lunar probe can effectively address the limitations of existing test platforms in environmental simulation, system evaluation, and data fusion. This has significant theoretical and practical value for improving the probe's autonomous decision-making capabilities and optimizing path planning algorithms. Summary of the Invention

[0006] To address the problems in existing technologies where lunar probe path planning function testing is often conducted late and cannot be performed on actual sites, and where algorithm defects are difficult to detect in a timely manner, this invention aims to provide a testing system and method for autonomous path planning functions of lunar probes. Through logic simulation and closed-loop verification, the path planning algorithm and related modules can be jointly tested and evaluated in the early stages of R&D, thereby reducing R&D costs, shortening the verification cycle, and improving system reliability.

[0007] To achieve the above objectives, this invention discloses a testing system for the autonomous path planning function of a lunar probe, comprising: Lunar Scene Simulation Module: Used to construct simulations of the lunar environment, including the simulation of terrain, obstacles, and lighting factors, providing realistic test scenarios for the probe's autonomous path planning; Running in ROS, it is responsible for loading and updating the lunar surface terrain digital model (including undulating terrain, obstacle locations and physical properties) in real time, and providing the current environmental information topic (e.g. " / environment / terrain", message type is a custom terrain information message) to the dynamics module and sensor module. Dynamics module: Used to simulate the motion state of a lunar probe, including linear and angular motion, and to simulate the probe's motion behavior in the lunar environment; It receives control commands (such as speed and steering) from the controller module and terrain information published by the lunar scene simulation module. It performs numerical calculations using the dynamics and kinematics model of the probe chassis and outputs millimeter-level information such as position, attitude and wheel contact status. It publishes this information to the sensor module through the ROS topic " / dynamics / state". Sensor module: Used to simulate sensor data input from the detector, including cameras, lidar, and inertial measurement units (IMU), and to generate sensor data with noise interference; Subscribe to the real motion state published by the dynamics module and the terrain information provided by the lunar scene simulation module. Generate noisy simulated sensor data according to the preset sensor parameters (including lidar point cloud model, visual camera calibration model, inertial measurement unit noise model, etc.), and send it to the controller module and data acquisition and storage module through ROS topics (such as " / sensor / lidar_scan", " / sensor / camera_image", " / sensor / imu"). Controller module: Based on environmental information such as the current position of the lunar probe, the target position, and the lunar surface terrain and obstacles, it plans the optimal driving path; As the subject under test, it subscribes to the environmental perception data output by the sensor module, runs the path planning algorithm to be verified (such as global / local planning algorithms based on A*, D* Lite or sampling method), generates motion control commands (speed, steering angle, etc.) in real time, and sends them to the dynamics module through the ROS topic " / planning / control_cmd". At the same time, it publishes the planned path (such as path point sequence) and planning decision log to the data acquisition and storage module in the form of messages. Data acquisition and storage module: used to collect detector status data, path planning results and environmental simulation data in real time, and store them in the database; During the test, the running status of each node is recorded in real time. All key topics of the lunar scene simulation module, dynamics module, sensor module and controller module are subscribed to. Simulation data is recorded in real time and stored in the database. After data acquisition, it needs to be stored in the database. Relational databases MySQL and PostgreSQL are used to store different types of data. Data storage adopts data backup and recovery mechanisms to prevent data loss.

[0008] Test and evaluation module: Used to compare the path planning results of the lunar probe with the expected path, and evaluate its autonomous path planning performance, including adaptability, efficiency and accuracy; After the test, the test evaluation module evaluates and compares the detector's performance in performing path planning tasks to ensure that the detector can meet the designed performance standards.

[0009] The aforementioned testing and evaluation module systematically evaluates multiple dimensions of the probe's path planning, such as adaptability, efficiency, and accuracy, helping developers and researchers identify and optimize path planning algorithms and ensuring that the lunar probe can autonomously, effectively, and safely complete its mission in the complex lunar environment.

[0010] The evaluation of lunar probe path planning mainly includes the following aspects: Environmental change response: Test whether the probe can sense changes in the lunar environment (such as sudden obstacles, terrain changes, etc.) and make effective path adjustments; Real-time decision-making efficiency: When facing unknown or uncertain environments, can the detector adjust its path planning strategy in real time to ensure the smooth progress of the mission? Dynamic path update efficiency: This test assesses whether the detector can update its path in real time during operation to cope with unexpected events or environmental changes. Algorithm computational efficiency: Whether the path planning algorithm can plan a feasible path within a reasonable time (e.g., within a few seconds or minutes) to ensure that the detector can perform tasks in real time during actual operation; Path length and time efficiency: In a given environment, is the path planned by the detector the shortest and most efficient, avoiding unnecessary detours and too many waypoints, thereby saving energy and reducing travel time; Resource utilization: Does the path planning reasonably consider the resource constraints of the probe (such as battery power, maximum speed, etc.) to ensure that the mission is completed with maximum resource utilization?

[0011] Under the coordination of ROS Master, the above modules form an end-to-end closed-loop logic. After the test-driven module initializes the scene, the lunar scene simulation module continuously updates the terrain and notifies the dynamics module; the dynamics module generates real motion states and feeds them to the sensor simulation; the sensor simulation outputs perception data for use by the path planning algorithm; the controller module calculates control commands and returns them to the dynamics module; the data acquisition and storage module records and evaluates the planning performance throughout the process.

[0012] The testing and evaluation module performs calculations based on several defined evaluation metrics. These metrics quantify the detector's performance during the path planning process and provide a basis for improvement.

[0013] By conducting autonomous path planning function tests on lunar probes, high-fidelity, multi-scenario, and scalable path planning function tests can be carried out without relying on real-world environments.

[0014] Preferably, the lunar scene simulation module in the above system can simulate different lunar environmental conditions according to the set lunar terrain and environmental conditions, and provide real environmental data input for testing.

[0015] Preferably, the dynamics module in the above system is a precise dynamics module to simulate the motion state of the lunar probe under different environmental conditions, ensuring the effectiveness and accuracy of the path planning system.

[0016] Preferably, the sensor module in the above system generates sensor data with noise interference by simulating the sensors on the lunar probe, so as to simulate various sensor errors and environmental interferences encountered in actual missions.

[0017] Preferably, the data acquisition and storage module in the above system can perform subsequent data analysis and backtracking by storing detector status information, path planning data and environmental condition data in real time, and supports long-term test processes.

[0018] Preferably, the test and evaluation module in the above system analyzes the autonomous path planning effect of the probe in the complex lunar environment by comparing it with the preset expected path, and provides a detailed evaluation report.

[0019] This invention also discloses a method for testing the autonomous path planning function of a lunar probe using the test system described above (hereinafter referred to as the "test method"), comprising the following steps: S1. Establish a digital model of the lunar environment: Using terrain data acquisition and preprocessing tools, generate a three-dimensional mesh map (DEM) containing lunar surface elevation, obstacle distribution and physical attributes, and convert the model into a message format that ROS can recognize and load it into the lunar scene simulation module. S2. Construct the detector dynamics module: Based on the detector chassis design parameters (such as center of mass position, tire material, reducer transmission ratio, etc.), establish kinematic equations and dynamic simulation models, write the corresponding ROS node code and deploy it in the dynamics module; S3. Configure sensor simulation parameters: Set the lidar scanning angle range and resolution, camera intrinsic and extrinsic parameters and frame rate, and the noise characteristics of the inertial measurement unit (IMU). Write the above parameters into the ROS parameter server, and the sensor module reads and initializes the simulation model. S4. Deploy path planning algorithm node: Encapsulate the path planning algorithm to be tested as a ROS node, configure global / local map subscription topics (obtained from the lunar scene simulation module or sensor module) and control command publishing topics; set planning frequency, heuristic function or cost function parameters; S5. Initialize the test-driven process: Start ROS Master and start the lunar scene simulation module, dynamics module, sensor module, controller module, data acquisition and storage module and test evaluation module in sequence. The test-driven module will send the test scene identifier, simulation time step, planning termination condition and performance index threshold in sequence through ROS service calls. S6. Execute automated testing: The test driver module sends a "start test" control command to the controller module. All modules run in parallel under the ROS communication framework. The lunar scene simulation module continuously publishes terrain information, the dynamics module simulates motion state, the sensor module outputs sensing data, the controller module calculates and issues control commands in real time based on the simulated sensing information, and the dynamics module updates the state, forming a closed loop. At the same time, the data acquisition and storage module records and stores data such as path trajectory, time cost, obstacle avoidance events, and positioning error in real time during the test. S7. Evaluation and Feedback: When the detector's simulated trajectory reaches the preset target point or meets the termination conditions, the test-driven module issues a "stop" command to each functional module. The test evaluation module calculates environmental change response, real-time decision-making efficiency, dynamic path update efficiency, algorithm calculation efficiency, path length and time efficiency, resource utilization, etc., and automatically determines whether the test passes based on the preset performance thresholds. The test evaluation module evaluates the path planning results and outputs a test report. If a planning anomaly is found (such as getting stuck in an infinite loop or failing to bypass obstacles), an anomaly log is recorded and a visual report is generated to notify the development team to improve the algorithm. S8. Iterative Optimization: Based on the data analysis results, designers can adjust the path planning algorithm parameters (such as cost function weights, heuristic search strategies, etc.) or modify the configuration of the dynamics / sensor model, and then execute steps S1-S7 again for closed-loop verification until the performance requirements are met.

[0020] Preferably, the test report in the above method provides optimization suggestions based on the path planning effect of the probe under different lunar surface environments, so as to improve the accuracy and efficiency of the probe's path planning under complex conditions.

[0021] By adopting the above technical solution, the present invention can achieve the following technical effects: This invention enables early logic verification and full-process closed-loop testing of the lunar probe path planning function. It can efficiently discover algorithm defects and evaluate performance bottlenecks without the need for a physical prototype. Furthermore, the distributed ROS architecture ensures the scalability and reusability of each functional module, greatly improving R&D efficiency and reducing testing risks.

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0023] Figure 1 This is a diagram showing the composition of the test system for autonomous path planning function of a lunar probe in this invention. Figure 2 This is a roadmap for implementing the autonomous path planning function test system for lunar probes in this invention; Figure 3 This is a schematic diagram of the scene construction of the lunar scene simulation module of the test system in this invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0026] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: This invention provides a testing system and method for autonomous path planning functions of lunar probes, aiming to solve problems such as the post-testing of existing path planning functions, lack of early verification methods, and difficulty in timely detection of algorithm defects. By introducing logic simulation and closed-loop verification mechanisms, this invention can perform integrated testing and performance evaluation of the path planning algorithm with sensing, control, and dynamics modules in the early stages of research and development, improving verification efficiency and system reliability.

[0027] Combination Figure 1-2 As can be seen, the system is built on a ROS distributed architecture and mainly includes a lunar scene simulation module, a dynamics module, a sensor module, a controller module, a data acquisition and storage module, and a testing and evaluation module. Each of these modules is deployed and runs as an independent ROS node, and the ROS message communication mechanism (Topic) and service mechanism (Service) are used to realize data interaction and control coordination between modules, forming a complete closed-loop testing process.

[0028] Combination Figure 3As can be seen, the lunar surface scene simulation module establishes a digital terrain model of the lunar surface. It acquires lunar surface elevation, obstacle distribution, and physical property data through terrain acquisition and processing tools, generates a 3D mesh map (DEM), and converts it into a ROS-readable data format for loading by the lunar surface scene simulation module. This constructs a realistic virtual lunar environment, including typical lunar landforms such as lunar rocks, craters, slopes, and dust areas, supporting parametric adjustments to terrain distribution, density, and roughness. This module runs on a high-performance simulation platform (such as Gazebo or Unreal Engine) and provides ROS-compatible data publishing, continuously publishing terrain information to the dynamics and sensor modules via the " / environment / terrain" topic to ensure environmental consistency.

[0029] The dynamics module simulates the dynamic response of a lunar probe, based on realistic mass distribution and six-degree-of-freedom dynamic modeling, considering lunar gravity, wheel-Earth interaction, and slippage effects. Deployed on a real-time computing node, this module updates the probe's position and attitude information in real time by subscribing to sensor data and path commands. It performs numerical simulations based on the probe's chassis kinematics and dynamics module, generating real-time information on the probe's position, attitude, and wheel states, which is then published via the topic " / dynamics / state" for use by subsequent modules.

[0030] The sensor module simulates various sensors carried by a lunar probe, including lidar, stereo camera, and IMU. Sensor data is published via ROS topics, providing sensory input for the path planning algorithm and state estimation modules. This module outputs noisy simulated sensor data according to set parameters, which is transmitted to the controller module and data acquisition module via topics such as " / sensor / lidar_scan", " / sensor / camera_image", and " / sensor / imu".

[0031] The test-driven module manages the test process, including loading test scenario configurations, issuing test control commands, and monitoring the running status of each module in real time. During the test initialization phase, the driver module loads terrain data, obstacle distribution, sensor parameters, algorithm configurations, and preset performance indicators through the ROS parameter server or service interface. During the test execution, it checks for node anomalies (such as planning timeouts, module crashes, etc.) and triggers interruptions if necessary.

[0032] The data acquisition and storage module is used to collect data output from each module in real time, including sensor output, detector status information, path point sequences, obstacle avoidance behavior, etc., and store them in a database or ROS bag file for subsequent analysis. It supports the automatic annotation of key events in path execution, such as obstacle avoidance points and failed paths.

[0033] To enable data acquisition modules to interact with other modules (such as controller modules, sensor modules, and control systems), the modules should provide standardized interfaces. The following interfaces can be designed: Data acquisition interface: Provides methods for acquiring real-time data from different sensors.

[0034] collect_sensor_data(sensor_type: str, timestamp: float) ->Dict sensor_type: Data type (e.g., LiDAR, IMU, Camera).

[0035] timestamp: The timestamp of data collection.

[0036] Data storage interface: Provides methods to store the collected data into the database.

[0037] store_data(data: Dict) data: Includes collected sensor data (raw data from IMU and LiDAR) and path planning data (such as planned paths, waypoints, and parameters of the planning algorithm).

[0038] Real-time data update interface: Provides methods to obtain and update data in real time.

[0039] update_data(data_id: int, new_data: Dict) data_id: A unique identifier (ID) used to specify the data that needs to be updated. Each piece of data in the system typically has a unique ID to ensure that the data can be accurately located and updated.

[0040] new_data: A dictionary containing the data that needs to be updated. The data in new_data will overwrite or update the previously stored old data.

[0041] The testing and evaluation module is used to analyze and evaluate the effectiveness of path planning, including indicators such as response to environmental changes, real-time decision-making efficiency, dynamic path update efficiency, algorithm computation efficiency, path length and time efficiency, and resource utilization. The system supports comparative testing of multiple algorithms, automatically generates evaluation reports and performance curves, and assists developers in optimizing algorithm parameters.

[0042] During implementation, users set the initial position and target point through the test control interface, and call the specified path planning algorithm (such as A*, RRT, D*, etc.). The system automatically starts the simulation process. The sensor module provides real-time environmental perception, and the path planning algorithm calculates the path and sends it to the dynamics module for execution. The execution results are recorded in real time by the data module, and finally, the evaluation module generates a test report.

[0043] Under the coordination of ROS Master, each module forms an end-to-end closed-loop logical flow: after the test-driven module completes scene initialization, the lunar scene simulation module provides terrain information in real time; the dynamics module calculates the motion state according to the planning commands; the sensor module generates sensing data; the controller module calculates the path in real time based on the sensing data; and all process data is recorded and analyzed by the data acquisition module. This structure supports high-fidelity, multi-scene, and scalable path planning function testing without a physical probe.

[0044] Furthermore, to improve testing efficiency, the system of this invention supports batch task queue operation and multi-algorithm comparison experiments; at the same time, each module supports Docker containerized deployment, which facilitates cross-platform testing and version control.

[0045] Through the above system and method, this invention realizes early simulation verification and closed-loop testing of path planning function, with high fidelity, high degree of automation and scalability. It can efficiently discover the defects and performance bottlenecks of planning algorithms without relying on physical detectors, and significantly improve the development efficiency and reliability of path planning system.

[0046] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A test system for autonomous path planning function of a lunar surface probe, characterized in that, The testing system includes: Lunar Scene Simulation Module: Used to construct simulations of the lunar environment, including the simulation of terrain, obstacles, and lighting factors, providing realistic test scenarios for the probe's autonomous path planning; Dynamics module: Used to simulate the motion state of a lunar probe, including linear and angular motion, and to simulate the probe's motion behavior in the lunar environment; Sensor module: Used to simulate sensor data input from the detector, including cameras, lidar, and inertial measurement units (IMU), and to generate sensor data with noise interference; Controller module: Based on environmental information such as the current position of the lunar probe, the target position, and the lunar surface terrain and obstacles, it plans the optimal driving path; Data acquisition and storage module: used to collect detector status data, path planning results and environmental simulation data in real time, and store them in the database; Test and evaluation module: Used to compare the path planning results of the lunar probe with the expected path, and evaluate its autonomous path planning performance, including adaptability, efficiency and accuracy; Each of the above modules operates as an independent ROS node in a distributed architecture, forming a closed-loop data flow through ROS topics and services. Specifically, the lunar scene simulation module provides current environmental information topics to the dynamics module and the sensor module. The dynamics module receives control commands from the controller module and terrain information published by the lunar scene simulation module, performs numerical calculations using the probe chassis dynamics and kinematics model, and outputs millimeter-level position, attitude, and wheel contact state information, which is then published to the sensor module via ROS topics. The sensor module subscribes to the real motion states published by the dynamics module and the terrain information provided by the lunar scene simulation module, and generates noisy data according to preset sensor parameters (including the LiDAR point cloud model, visual camera calibration model, and inertial measurement unit noise model). The simulated acoustic sensor data is sent to the controller module and the data acquisition and storage module via ROS topics. The controller module, as the subject under test, subscribes to the environmental perception data output by the sensor module, runs the path planning algorithm to be verified, generates motion control commands in real time, and sends them to the dynamics module via ROS topics. The planned path and planning decision log are published to the data acquisition and storage module in the form of messages. During the test, the data acquisition and storage module records the running status of each node in real time, subscribes to all key topics of the lunar scene simulation module, the dynamics module, the sensor module, and the controller module, records simulation data in real time, and stores it in the database. After the test is completed, the test evaluation module evaluates and compares the performance of the probe when performing the path planning task to ensure that the probe can meet the design performance standards.

2. The testing system according to claim 1, characterized in that, The lunar scene simulation module can simulate different lunar environment conditions based on the set lunar terrain and environmental conditions, and provide real environmental data input for testing.

3. The testing system according to claim 1, characterized in that, The aforementioned dynamics module employs precise dynamics to simulate the motion of a lunar probe under different environmental conditions, ensuring the effectiveness and accuracy of the path planning system.

4. The testing system according to claim 1, characterized in that, The sensor module generates sensor data with noise interference by simulating the sensors on the lunar probe, in order to simulate various sensor errors and environmental interferences encountered in actual missions.

5. The testing system according to claim 1, characterized in that, The data acquisition and storage module stores detector status information, path planning data, and environmental condition data in real time, enabling subsequent data analysis and backtracking, and supporting long-term testing processes.

6. The testing system according to claim 1, characterized in that, The test and evaluation module compares the probe's autonomous path planning performance with the preset expected path to analyze the effectiveness of the probe in the complex lunar environment and provides a detailed evaluation report.

7. A test method based on the test system according to any one of claims 1 to 6, characterized in that, Includes the following steps: S1. Establish a digital model of the lunar environment: Using terrain data acquisition and preprocessing tools, generate a three-dimensional mesh map (DEM) containing lunar surface elevation, obstacle distribution and physical attributes, and convert the model into a message format that ROS can recognize, and load it into the lunar scene simulation module. S2. Construct the detector dynamics module: Based on the detector chassis design parameters, establish kinematic equations and dynamic simulation models, write the corresponding ROS node code, and deploy it in the dynamics module. S3. Configure sensor simulation parameters: Set the lidar scanning angle range and resolution, camera intrinsic and extrinsic parameters and frame rate, and the noise characteristics of the inertial measurement unit (IMU). Write the above parameters into the ROS parameter server, and the sensor module reads and initializes the simulation model. S4. Deploy path planning algorithm node: Encapsulate the path planning algorithm to be tested as a ROS node, configure global or local map subscription topics (obtained from the lunar scene simulation module or the sensor module) and control command publishing topics, and set the planning frequency, heuristic function or cost function parameters. S5. Initialize the test-driven process: Start ROS Master and sequentially start the lunar scene simulation module, dynamics module, sensor module, controller module, data acquisition and storage module, and test evaluation module. The test-driven module sequentially issues the test scene identifier, simulation time step, planning termination condition, and performance index threshold through ROS service calls. S6. Execute automated testing: The test driver module sends a "start test" control command to the controller module. Each module runs in parallel under the ROS communication framework. The lunar scene simulation module continuously publishes terrain information, the dynamics module simulates motion state, the sensor module outputs sensing data, the controller module calculates and issues control commands in real time based on the simulated sensing information, the dynamics module updates the state to form a closed loop, and the data acquisition and storage module records and stores the path trajectory, time overhead, obstacle avoidance events, and positioning error data in real time during the test. S7. Evaluation and Feedback: When the detector's simulated trajectory reaches the preset target point or meets the termination condition, the test drive module issues a "stop" command to each functional module. The test evaluation module calculates the response to environmental changes, real-time decision-making efficiency, dynamic path update efficiency, algorithm calculation efficiency, path length and time efficiency, and resource utilization. Based on the preset performance threshold, it automatically determines whether the test passes. The test evaluation module evaluates the path planning results and outputs a test report. If an anomaly is detected in the planning (such as getting stuck in an infinite loop or failing to bypass obstacles), an anomaly log is recorded and a visual report is generated to notify the development team to improve the algorithm. S8. Iterative Optimization: Based on the data analysis results, designers can adjust the path planning algorithm parameters or modify the configuration of the dynamics or sensor model, and then execute steps S1-S7 again for closed-loop verification until the performance requirements are met.

8. The test method according to claim 7, characterized in that, The test report provides optimization suggestions based on the probe's path planning performance under different lunar surface environments, in order to improve the accuracy and efficiency of the probe's path planning under complex conditions.

Images