Research on Path Planning Method of Pipe Descaling Robot

Aiming at the existing pipeline robots' problems of low trajectory tracking accuracy, poor adaptability to complex pipeline environments, and many redundant inflection points in the path, we propose an improved A-star algorithm based on the safe distance constraints and the principle of instant line of sight, and combined with the bidirectional slider optimization strategy, which reduces the path length and computational complexity. By modeling the pipeline environment, the designed algorithm is simulated and analyzed and compared with other algorithms for path planning in pipelines. The results show that the path length planned by the proposed method is about 8.1% shorter than Dijkstra algorithm and about 1.1% shorter than JPS algorithm. The number of turning points is only 62.5% of other algorithms. The cumulative turning angle is reduced by 53.2% compared to Dijkstra algorithm and 13.9% compared to JPS algorithm. The experimental results verify the advantages of the optimized algorithm.