Research on path planning of intelligent maintenance robotic arm for distribution lines under complex environment
Maintenance of distribution lines primarily relies on manually operated robotic arms, which lack intelligence and pose safety risks of collisions in complex distribution network environments. In this paper, the performance of sampling-based algorithms such as Rapidly-exploring Random Trees (RRT), RR...
Gespeichert in:
Veröffentlicht in: | Computers & electrical engineering 2024-12, Vol.120, p.109711, Article 109711 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Maintenance of distribution lines primarily relies on manually operated robotic arms, which lack intelligence and pose safety risks of collisions in complex distribution network environments. In this paper, the performance of sampling-based algorithms such as Rapidly-exploring Random Trees (RRT), RRT*, and RRT-Connect for path planning in complex 10 kV distribution environments are thoroughly investigated. Firstly, the performance differences between the algorithms in a simple 2D experimental environment and a narrow 3D environment are compared, focusing on the metrics of path planning time, path length and number of path nodes. In addition, tests are conducted on a constructed experimental platform of 10 kV distribution line, and the results show that the path planning success rate of RRT-Connect algorithm reaches 100%, which is significantly higher than 90% and 60% of RRT and RRT*. RRT-Connect reduces the average planning time to 155 ms, saving 64% and 96% time compared to RRT and RRT*, respectively, which demonstrates excellent reliability and operational efficiency. These results validate the feasibility of the RRT-Connect algorithm in the practical application of intelligent maintenance robotic arms for distribution lines, providing substantial technical support for improving the safety and intelligence of distribution network maintenance.
[Display omitted]
•The use of stable and efficient path planning algorithms in the field of distribution network maintenance robotic arms can effectively solve the problems of low intelligence level and potential safety hazards caused by collisions.•The validation experiments on 10 kV distribution line experimental platform demonstrate a 100% planning success rate for RRT-Connect, significantly outperforming RRT and RRT* with success rates of 60% and 90%, respectively.•Although the average planning path length of RRT-Connect, at 50.8 cm, is 14% longer than RRT*, it compensates with a remarkable speed advantage.•The average planning time for RRT-Connect is 155 ms, which is 64% and 96% faster than RRT and RRT*, respectively. This reduction in planning time makes RRT-Connect suitable for distribution line maintenance where response speed is crucial. |
---|---|
ISSN: | 0045-7906 |
DOI: | 10.1016/j.compeleceng.2024.109711 |