← Paper Digest

Modeling the Frictional Contact of Continuum Robots with Environments through Linear Complementarity Problem

Jia Shen, Junhyoung Ha, Yue Chen

November 2026

TL;DR A quasi-static model of continuum robots rubbing against their surroundings solves contact forces, sliding and robot shape together as a linear complementarity problem, so it captures sticking, sliding and hysteresis. On a 287.9 mm tendon-driven robot pressed against a wall, the maximum tip error drops from 45.4 mm (frictionless model) to 5.85 mm.

What they do

Continuum robots such as concentric-tube robots and tendon-driven catheters often rub against their surroundings, and that friction can move the tip by centimeters. Earlier models assume the robot always slides opposite to the actuation, so they predict only the extreme shapes and miss sticking and the hysteresis when actuation reverses. This paper models friction with the environment through complementarity constraints, so each contact can stick, slide and switch between the two over an actuation history.

How it works

  • The idea comes from rigid-body contact, but instead of a mass-based dynamic formulation they write a quasi-static one built on the robot’s actuation Jacobian and compliance matrix from a Cosserat rod model.
  • At each small step one LCP gives the normal and friction forces (friction cone approximated by a polygon) and the sliding at every contact, and the shape is then updated.
  • Contacts are found from the signed distance to the environment along the discretized rod, keeping one point per contact and treating surface and corner contacts differently.
  • They prove the LCP has a solution in stable configurations and that without friction the model matches their earlier frictionless contact model. Code is public.

Results

  • One step takes at most 11.06 ms in MATLAB on a CPU (over 90 Hz) in four simulated scenarios; the LCP itself takes under 1.2 ms.
  • Nitinol tube against a rotating plane: mean body error 2.69 mm and maximum 6.64 mm (about 1.3% and 3.3% of tube length). Inside a rotating pipe: mean body error 0.55 mm and maximum tip error 1.26 mm, where the frictionless model predicts no tip deflection.
  • Tendon-driven catheter with a notched needle against an aluminum wall: maximum tip error 5.85 mm and bending-angle error 4.6°, against 45.4 mm and 60.6° for the frictionless model.

Limitations

Stick-slip is not modeled (the friction coefficient is constant), and it is the main error source. The model is quasi-static, assumes point contacts and needs small steps because it linearizes once per step. Friction between the tubes of a concentric-tube robot is not covered. Friction coefficients are fitted per setup, and the sub-millimeter errors in the pipe tests are below the 1.11 mm error of the fiber shape sensor used as reference.

Figures

From Figure 2: Frictional contact between a continuum robot and a rigid environment. (a) The Coulomb friction cone ap...
From Figure 2: Frictional contact between a continuum robot and a rigid environment. (a) The Coulomb friction cone approximated by a polygon. (b) Over one time step the obstacle moves and the robot is actuated; the contact force inside the friction cone and the relative sliding at the contact are solved together as an LCP.
From Figure 11: The physical setup. A precurved nitinol tube with an embedded fiber shape sensor is fixed at its base...
From Figure 11: The physical setup. A precurved nitinol tube with an embedded fiber shape sensor is fixed at its base and bent by friction as the environment rotates: (a) pressed by a plane while held inside a ring, (b) the tube, (c) inside a rotating straight pipe.
From Figure 18: A tendon-driven continuum robot against a rigid wall, with model predictions overlaid. (a) The base p...
From Figure 18: A tendon-driven continuum robot against a rigid wall, with model predictions overlaid. (a) The base pushes forward and compresses the robot, (b) the tip keeps contact while the catheter bends, (c) contact breaks as it bends further. (d, e) Tip x-position and bending angle over time: measurement, friction model and frictionless model.
Abstract

Accurate friction modeling in continuum robots is essential for precise motion control and path planning. However, current approaches make predictions only for the mechanism's internal friction that is limited by the boundary forces of the Coulomb friction cone, which fail to characterize the static friction and motion hysteresis. In this paper, we propose a novel friction model based on complementarity constraints under the Coulomb friction assumption. This model predicts various friction effects caused by environmental interactions, including both static and sliding frictions and hysteresis behavior, as well as the corresponding robot deformation induced by these interactions, which are crucial for the practical motion planning and control of continuum robots. Additionally, real-time implementation of this model can be achieved, meeting the demand for computational efficiency in robotic applications. Experimental validations demonstrated that this model is able to reduce the maximum tip position error from 45.4 mm in the frictionless model to 5.85 mm in our friction-included model with a 287.9 mm-long tendon-driven continuum robot.

Summaries are written automatically from each paper's text. Figures and videos come from the paper and its project page; all credit goes to the authors.