Feature Selection Based on Reinforcement Learning and Hazard State Classification for Magnetic Adhesion Wall-Climbing Robots

Abstract: Magnetic adhesion tracked wall-climbing robots face potential risks of overturning during high-altitude operations, making their stability crucial for ensuring safety. This study presents a dynamic feature selection method based on Proximal Policy Optimization (PPO) reinforcement learning, combined with typical machine learning models, aimed at improving the classification accuracy of hazardous states under complex operating conditions. Firstly, this work innovatively employs a fiber rod-based MEMS attitude sensor to collect vibration data from the robot and extract high-dimensional feature vectors in both time and frequency domains. Then, a reinforcement learning model is used to dynamically select the optimal feature subset, reducing feature redundancy and enhancing classification accuracy. Finally, a CNN-LSTM deep learning model is employed for classification and recognition. Experimental results demonstrate that the proposed method significantly improves the robot's ability to assess hazardous states across various operational scenarios, providing reliable technical support for robotic safety monitoring. Keywords: Magnetic Adhesion Wall-Climbing Robot, MEMS Sensor, Hazard State Evaluation, Reinforcement Learning, Feature Selection, Deep Learning 1. Introduction Magnetic adhesion tracked wall-climbing robots are designed specifically for vertical or inclined surfaces, enabling them to effectively counteract gravity and perform a variety of tasks [1], such as inspection, welding, and cleaning in high-altitude environments [2-5]. These robots have broad application prospects, particularly in dangerous high-altitude operations, where they can significantly improve work efficiency and ensure the safety of operators [6]. However, as the robot moves along the wall, the overturning torque generated by its weight and load may cause it to flip backward, affecting its stability and posing potential safety risks [7].