Hybrid Impedance Control of Cable-Driven Parallel Robots Using Adaptive Friction Compensation

Cable-driven parallel robots (CDPRs) are a new type of parallel robot using several cables to drive a mobile platform, and perform better than rigid parallel robots in certain cases. However, CDPRs suffer from friction disturbances due to their complicated flexible mechanism, which will decrease the control accuracy and bring challenge to contact-rich tasks. To handle these problems while achieving impedance control, a hybrid impedance control method is proposed for realizing robust impedance control and friction compensation for CDPRs simultaneously. Different from ordinary modifications of sliding mode control methods with complex sliding function, the innovation of our method is to propose a novel hybrid impedance controller with an adaptive radial basis function neural network (RBFNN) applied to the sliding mode impedance control (SMIC) framework for friction compensation. With such consideration, the friction in the CDPRs is precisely compensated, and the chattering phenomenon of the sliding mode controller is significantly reduced. Eventually, our method serves as a general impedance control scheme for CDPRs to maintain safety and accuracy in contact-rich tasks. According to comparative experiments, our method brings more performance improvement of 33.6% and 34.0%, respectively than the impedance controller based on fast terminal sliding mode synchronous control (FTSMC) for two CDPR configurations in two representative contact-rich scenarios.