Variable Curvature Path Planning for Robot-Assisted Flexible Needle Insertion Based on Improved Bi-RRT Algorithm

Flexible-needle percutaneous puncture is a widely used medical technique in clinical practice and is known for its advantages of minimal patient, fast recovery, and ease of operation. However, ensuring accuracy and avoiding obstacles during the puncture process are challenging because of the non-holonomic motion of the needle tip and the presence of obstacles and sensitive organs. To address this issue, this study proposed a novel path planning algorithm for robot-assisted flexible needle punctures. The algorithm is based on a bidirectional rapidly exploring random tree and incorporates target-oriented and rapid-search strategies. This provides a feasible approach for the clinical application of flexible-needle puncture, outperforming commonly used algorithms in terms of computational speed, path representation, and search capabilities. Additionally, this study applies a duty-cycling control strategy for a flexible needle, enabling proportional control of the curvature of the needle trajectory during soft tissue insertion. The effectiveness of the proposed path planning algorithm was validated through experiments using a duty-cycling strategy, and an experimental error within 4 mm was deemed adequate for most clinical needle-insertion applications.