Intelligence Demand-Driven Automatic Path Planning Study for Substation Inspection Robots

• Published in: 2024 5th International Conference on Artificial Intelligence and Computer Engineering (ICAICE) pp. 983 - 986
• Main Authors: Wang, Hongkai, Feng, Jun, Zhao, Shuai, Wang, Yidan, Song, Haomin
• Format: Conference Proceeding
• Language: English
• Published: IEEE 08.11.2024
• Subjects: Accuracy; automatic path planning; Inspection; intelligence requirements; Libraries; Optimization; Path planning; Real-time systems; Reliability; robot navigation; Robots; Safety; substation inspection; Substations
• DOI: 10.1109/ICAICE63571.2024.10864119

[Purpose/Significance] The problems of insufficient path planning accuracy and poor environmental adaptability during substation inspection are rooted in the lack of a clear and accurate description of intelligence requirements in complex environments, which leads to insufficient inspection efficiency and safety. This study aims to improve the intelligence level of automatic path planning for substation inspection robots and reduce the risk of power accidents through an intelligence demand-driven approach. [Methods/Process] This study emphasizes the key role of intelligence requirements in the whole process of substation inspection. By identifying the intelligence requirements in the substation environment, the inspection tasks are fine-grained, the substation contingency task module library is constructed, and the intelligence requirement association library for inspection operations and execution subjects is established. Based on this, an automatic route optimization strategy integrating A* algorithm and potential field approach is proposed to form a research framework to effectively respond to inspection tasks under the complex environment of substations. [RESULTS/CONCLUSIONS] The constructed fine-grained task module library and intelligence requirement association library are established before inspection and dynamically updated during inspection. The synergistic effect of these two libraries makes the identification of intelligence requirements in the inspection process faster and more objective, enhances the robot's precision when acquiring environmental information, and enhances the flexibility and reliability of route determination.