Design, optimization, and crosstalk analysis for robot joint torque sensor of shear beam structure.

This article focuses on the design, optimization, and experimental validation of a spoke-type joint torque sensor (JTS) based on the shear beam principle for collaborative robots. Using the Response Surface Method (RSM) and Multi-Objective Genetic Algorithm (MOGA), the sensor's structural parameters were optimized to balance sensitivity, stiffness, crosstalk resistance, and overload capacity. Experimental results demonstrate that the optimized sensor achieves high overload resistance (300% full scale), low measurement errors under axial, radial, and moment crosstalk loads, and meets engineering requirements for linearity, repeatability, and hysteresis. The shear beam design offers superior anti-crosstalk performance, especially against axial loads, compared to bending beam sensors, making it suitable for human–robot interaction applications in Industry 5.0 contexts.