A novel path generation method for robotic measurement with local pruning and collision-free adjustment

Abstract Robotic measurement is an essential task during downtime in the nuclear power underwater environment. However, the narrow operating space of the underwater nuclear environment brings about an enormous challenge about the efficiency and safety of robotic measurement. This paper proposes a novel path-generation method with local pruning and collision-free adjustment. The proposed method increases the distance between the path and obstacles and reduces the number of collision detections. This core idea is to calculate extreme value points and path feature points as well as the collision-free adjustment for a smooth path. Firstly, the path is divided into subintervals where interval’s endpoints are the extreme value points. Next, the path points are distinguished by path feature points in each subinterval. A two-level connection strategy is proposed to connect the path points among path feature points within the constraints of curvature and density. Then, the smooth path is adjusted to avoid obstacles. The use of extreme value points and path feature points contributes to increasing the efficiency of path generation and improving the safety of the robotic measurement. Simulations and practical experiments are carried out to validate the feasibility of the proposed method over another method. Path safety index (PSI) is defined to measure the distance between path and obstacles. The efficiency of path generation is improved by up to 101% and PSI is improved by up to 20.75%.