Temperature-Based Contact Sensing and Force Estimation for Discrete-Jointed Continuum Robots

Contact force is a critical feedback mechanism for ensuring safe robot-environment interactions. Integrating dedicated force sensors into continuum robots (CRs) is difficult due to strict size constraints. Consequently, estimating contact forces based on the robot’s shape/curvature feedback has gained widespread recognition. This involves solving an optimization problem that minimizes the discrepancy between predicted and actual shapes/curvatures. However, this optimization problem is typically high-dimensional, which often leads the optimization to converge to local minima. This article introduces a novel framework for external force estimation based on temperature sensing. The number of external forces, their application points and directions are uniquely determined from the temperature variation during CR-human contact. These identified parameters are incorporated as known inputs into the optimization problem, substantially reducing its dimensionality and consequently improving global convergence, estimation accuracy, and computational efficiency. Based on the CR shape obtained from electromagnetic sensors, force estimation under conditions of known and unknown contact information was compared through simulations and experiments. Results show that the proposed framework significantly improves global convergence compared with methods without contact information, while also greatly reducing computation time. Furthermore, the framework requires only a few compact temperature sensors, which do not increase the overall size of the CR.