A polymer-based compliant force/torque and displacement sensor with creep compensation

Abstract

Automation of assembly processes using robots remains challenging, requiring advanced capabilities in sensing for assembly state estimation and shock absorption to protect the robot from external forces. Traditional devices such as passive compliance devices and force/torque sensors provide only a subset of these functionalities. To integrate these capabilities into a single system, a polymer-based compliant force/torque and displacement sensor has been proposed. However, polymers classified as viscoelastic materials exhibit significant creep, degrading the sensing performance. Therefore, this paper addresses the development of a polymer-based compliance force/torque and displacement sensor with a creep compensation algorithm. The proposed sensor is made of viscoelastic materials to provide passive compliance, allowing it to adapt to external forces and protect the robot from shocks. Especially, it can achieve a remote center of compliance through a novel deformable structure. Moreover, it can measure the six-axis external forces and displacements generated by the passive compliance. In this process, the creep of the viscoelastic material is analyzed and compensated to improve the sensing performance. First, a stiffness analysis was conducted for the design of the sensor and finite element analysis was performed to verify that the sensor has a remote center of compliance. Then, a method for applying a creep compensation algorithm based on a viscoelastic model to the multi-axis force/torque and displacement sensor was proposed. The effectiveness of the creep compensation algorithm was evaluated through experiments, resulting in an 85.42 % reduction in creep error, a 63.52 % improvement in response time, and a 49.37 % reduction in hysteresis error. The proposed sensor can be utilized in various fields that require both flexibility and task state estimation such as robotic assembly, wearable robots, and medical robots. © 2025 Elsevier Ltd