Amended probabilistic roadmaps (A-PRM) for planning the trajectory of robotic surgery

Abstract

Trajectory planning is an essential aspect of research into the use of pliable needles for surgical processes. Sampling-based algorithms can generate trajectories and reach the target avoiding obstacles. However, the trajectories cannot match the physical constraints of injecting the pliable needle into human flesh, as the trajectories are not continuous. Aimed at solving this problem, an enhanced probabilistic roadmap (PRM) is used in this work. A PRM generates trajectories for surgeries that are minimally invasive and simultaneously guarantees the effectiveness and continuity of the trajectory. In this research work, the classical PRM method is enhanced by using a shape preserving piecewise cubic hermite interpolation (PCHIP) technique, used to generate smooth trajectories, which are important for navigating the curved path of the pliable needle in surgery. Trajectories that have been generated using the PRM satisfy direction constraints approach in terms of both source and target positions. As a result, the trajectories produced by the pliable needle are dynamically and geometrically feasible. Results of simulations performed show the validity of the algorithm implying that it can be efficiently used in trajectory planning of pliable needles in real-time surgical operations.