Improved Pulley Set-Based Force Sensing Mechanism Used in a Cable-Driven Exoskeleton Under Nonlinear Framework

Cable-driven exoskeletons are widely used for rehabilitation training and have advantage of being light-weight and low power. In human-machine tasks, the measurement of interactive force is important. Measuring the cable tension is a direct solution to obtain interactive force. For coupled controlled exoskeletons, the measurement of cable tension is challenging due to movement interference and friction. To further solve the abovementioned challenges, we propose a new scheme, which includes a force-sensing mechanism based on pulley sets, a drive efficiency observer-based terminal sliding mode controller, decoupled control, and safety strategy. The demonstration is carried out through modeling, simulation and prototype experiments to drive the exoskeleton for rehabilitation training and to accurately obtain the cable tension. The mean measurement error of the proposed method is shown to be 0.119 N, which takes only 1/3 of the required force. Compared to the traditional method, the proposed method is proved to be improved in the simulation with R2 : 98.6%, e RMSE : 82.38%, and the mean error: 91.78%. In the actual mirror training, R2 / e RMSE is 0.954/2.601 for flexion/ extension and 0.948/4.703 for supination/ pronation.