cs.ROSep 25, 2026

Anatomy-Aware Dexterity-Driven Design Optimization of Surgical Continuum Robots

Authors: Tony Qin, Peter Connor, Khoa Dang, Carter Hatch, Caleb Rucker, Robert J. Webster, Ron Alterovitz

Organizations: Department of Computer Science, University of North Carolina at Chapel Hill · Department of Mechanical Engineering, Vanderbilt University · Department of Mechanical, Aerospace, and Biomedical Engineering at The University of Tennessee, Knoxville

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Performing complex medical procedures with continuum robots requires careful selection of their geometric design parameters. The robot should have high dexterity in the specific anatomical environment of its procedure. This work presents a design optimization method that considers both dexterity and anatomy. We introduce the Reachable Volumetric Dexterous Solid Angle (RVDSA) metric as our objective, which measures the ability of a robot's end effector to reach the points in a goal volume from different directions via collision-free paths from a start configuration. We present a computationally efficient motion planner to compute this objective function for a given robotic design, and we use an asymptotically optimal simulated annealing optimizer to compute an optimized design. We applied our new method to optimize the design of a bimanual dexterous sheaths robot for performing procedures on cancerous polyps in colon anatomies, achieving a 78% higher RVDSA on average than optimizing for 3D voxel coverage alone.

Figures & tables

Fig. 1 : We define the RVDSA metric to evaluate the quality of the geometric design of a surgical continuum robot and apply it to the dexterous sheaths robot for procedures in the colon. (a) Colon anatomy from a CT scan and computed centerline. (b) Constant curvature backbones reaching target voxels in the colon anatomy from the computed colonoscope location depicted as a black dot. (c) The discretization of the goal volume G\mathcal{G} into voxels. Each polyp voxel has a discretized unit sphere representing the directional component of the end effector. (d) Discretization of the unit sphere S2\mathbb{S}^{2} into a subdivided icosahedron. Each triangle on the discretized unit sphere is marked green if it is reached at that voxel and direction. The RVDSA metric is computed from the percentage of triangles reached.
Fig. 2 : The five colon anatomies of our dataset with the location of each polyp circled in red.
Fig. 3 : (a) Random sampling. (b) Inverse kinematics sampling that brings the end effector to a goal voxel. (c) Self-motion sampling that reaches a goal voxel from many different directions.
Fig. 4 : Rendering of the bimanual dexterous sheaths robot in the colon reaching a cancerous polyp from many different directions. The design variables d1d_{1} to d3d_{3} are set at time of fabrication of the sheath’s constituent tubes and remain fixed throughout a procedure. During a procedure, the robot is actuated through its degrees of freedom q1q_{1} to q6q_{6} . Note that the tubes can be retracted so the full length of d1,d2,d3d_{1},d_{2},d_{3} may not be exposed.
Design VariableRange
d1:d_{1}: Outer pair length[0,50][0,50] mm
d2:d_{2}: Outer pair straight section length[0,50][0,50] mm
d3:d_{3}: Inner pair length[0,50][0,50] mm
TABLE I : Design Variables
TABLE II : Degrees of Freedom
Fig. 5 : The RVDSA is plotted as a function of run time for our method, Jacobian-only sampling, random-only sampling, and a voxel-optimized method. Each condition is run four times, with the mean and variance being displayed. The best RVDSA achieved by our method is 2.652.65 steradians, which is equivalent to a cone angle of 109.3109.3 degrees.
Fig. 6 : We compute the RVDSA of the best designs found across all runs for three different conditions. We display the discretized sphere coverage percentage for each individual triangle averaged over all polyp voxels and the two end effectors, after normalizing directions to the scope start frames.

Explore similar work

CardsList

  1. Kinematic Optimization of Phalanx Length Ratios in Robotic Hands Using Potential Dexterity

    Apr 22, 2026HyoJae Kang, Joonho Lee, Jeongdo Ahn +1Robotic Hand DesignMechanism Design

    Advanced Robotics Research Center, Korea Institute of Machinery & Materials (KIMM)

  2. Contact-Aware Planning and Control of Continuum Robots in Highly Constrained Environments

    Apr 17, 2026Aedan Mangan, Kehan Long, Ki Myung Brian Lee +3Constrained Motion PlanningContinuum Robot Control

    Department of Mechanical and Aerospace Engineering, University of California San Diego, San Diego, CA 92093, USA · SceniX Inc, New York, NY 10011, USA · Department of Electrical and Computer Engineering, University of California San Diego, San Diego, CA 92093, USA +1

  3. Capstan-driven Continuum Surgical Robot: Design, Modeling, and Perception

    Aug 13, 2026Gang Zhang, Yufu Qiu, Junyan Yan +3Continuum Robot ControlRobot State Estimation

    Department of Mechanical and Automation Engineering and T Stone Robotics Institute, The Chinese University of Hong Kong, Hong Kong · School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China · state Key Laboratory of Intelligent Manufacturing Equipment and Technology, Huazhong University of Science and Technology (HUST), Wuhan, China +1