Search papers, labs, and topics across Lattice.
This paper introduces a path planning method for surgical robots that operates in joint space by transforming the robot's position into a Riemannian manifold. An edge cost function is defined to optimize the path and reduce joint motion range, leveraging the non-concave nature of organs to facilitate gradient descent. Experiments show the proposed method reduces joint angle movement compared to path planning in position space.
Surgical robots can navigate more efficiently and with less joint strain by planning paths in joint space using Riemannian manifolds, outperforming traditional position-space methods.
Robotic surgery for minimally invasive surgery can reduce the surgeon's workload by autonomously guiding robotic forceps. Movement of the robot is restricted around a fixed insertion port. The robot often encounters angle limitations during operation. Also, the surface of the abdominal cavity is non-concave, making it computationally expensive to find the desired path.In this work, to solve these problems, we propose a method for path planning in joint space by transforming the position into a Riemannian manifold. An edge cost function is defined to search for a desired path in the joint space and reduce the range of motion of the joints. We found that the organ is mostly non-concave, making it easy to find the optimal path using gradient descent method. Experimental results demonstrated that the proposed method reduces the range of joint angle movement compared to calculations in position space.
