Sensory feedback performance improvement on RoboCab: An experimental approach to wire-driven parallel manipulator

This paper introduces a method to improve the feedback accuracy using multisensor data fusion. Flexibility in cables and unmodeled rotation of feeder pulley in the spatial workspace cause unreliability in the encoder data as well as sagging in cables. The purpose of this work is to provide an effective scheme to combine other sensors' information from multiple sources to increase the precision and reliability of feedback data. In this study, sensor level fusion algorithm is used to integrate and fuse together the camera and Inertial Measurement Unit (IMU) data. The main application of this method is in the implementation of various control methodologies where more accurate feedback data leads to an improved control performance. It is assumed that the cable mass is negligible and hence it insignificantly contributes to the robot dynamics. Experimental results show the merits of the proposed approach which will be discussed.