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

2025-03-22

AI Summary

To address the challenge of real-time overturning risk assessment for magnetic-adhesion tracked wall-climbing robots operating at height, this paper proposes a Proximal Policy Optimization (PPO)-driven integrated framework for dynamic feature selection and hazardous state classification. We innovatively employ a MEMS-based fiber-optic mast attitude sensor to acquire high-fidelity vibration and time-series pose features. A hybrid CNN-LSTM model is developed to fuse high-dimensional time-frequency domain features. Furthermore, a PPO-based dynamic feature subset selection mechanism is designed to suppress redundancy while enhancing discriminability. Experimental results across multiple operational scenarios demonstrate that the proposed method significantly improves hazardous state classification accuracy—achieving an average 9.7% gain over baseline models—while exhibiting strong robustness and real-time performance. This work provides a deployable technical foundation for safety monitoring of magnetic-adhesion climbing robots.

Technology Category

• Intelligent Robots: State Estimation
• Search and Optimization: Mixed Discrete/Continuous Search
• Natural Language Processing: Safety and Robustness

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

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.

Problem

• Improving hazardous state classification for wall-climbing robots
• Reducing feature redundancy using reinforcement learning
• Enhancing robot safety with dynamic feature selection

Innovation

• PPO reinforcement learning for dynamic feature selection
• Fiber rod-based MEMS sensor for vibration data
• CNN-LSTM model for hazardous state classification