Path Planning for Medical Robots – Approaches and Experimental Evaluation –

ath planning is a fundamental challenge in medical robotics, demanding precision and safety in complex environments. This paper reviews traditional and AI-based path planning approaches, focusing on their applicability in medical settings. A simulation-based experimental framework was developed, incorporating a UR5e robotic arm and NVIDIA Isaac Sim. To assess the framework’s capabilities, a path planning experiment was conducted comparing the RRT algorithm and the cuRobo motion planner by NVIDIA, with cuRobo achieving a 35.6 % improvement over the RRT algorithm. While the experiment was limited to a reduced scenario, the results illustrate the potential of the setup to evaluate key metrics such as computational efficiency, safety margins, and path optimality. The study highlights the strengths of the experimental framework and its components as a foundation for future, more complex investigations into path planning in medical environments.