Multi-source micro-friction identification for a class of cable-driven robots with passive backbone

• Main Authors: Tjahjowidodo, Tegoeh, Zhu, Ke, Dailey, Wayne, Burdet, Etienne, Campolo, Domenico
• Other Authors: School of Mechanical and Aerospace Engineering
• Format: Article
• Language: English
• Published: 2017
• Subjects: Transparent haptic interface; Cable-driven robot with passive backbone
• Online Access: https://hdl.handle.net/10356/83111; http://hdl.handle.net/10220/42439
• Institution: Nanyang Technological University

This paper analyses the dynamics of cable-driven robots with a passive backbone and develops techniques for their dynamic identification, which are tested on the H-Man, a planar cabled differential transmission robot for haptic interaction. The mechanism is optimized for human–robot interaction by accounting for the cost-benefit-ratio of the system, specifically by eliminating the necessity of an external force sensor to reduce the overall cost. As a consequence, this requires an effective dynamic model for accurate force feedback applications which include friction behavior in the system. We first consider the significance of friction in both the actuator and backbone spaces. Subsequently, we study the required complexity of the stiction model for the application. Different models representing different levels of complexity are investigated, ranging from the conventional approach of Coulomb to an advanced model which includes hysteresis. The results demonstrate each model's ability to capture the dynamic behavior of the system. In general, it is concluded that there is a trade-off between model accuracy and the model cost.