Intrinsic Tactile Sensing for the Optimization of Force Distribution in a Pipe Crawling Robot

This paper describes a tactile sensing system based on a force/torque sensor for the feet of a pipe crawling robot. Such a sensing system is needed for better optimization of force and joint load distribution and a safer avoidance of the risk of foot slippage. While conventional tactile sensing devices typically provide information concerning the spatial distribution of normal pressures, the intrinsic contact sensing system presented in this text only measures the three components of the contact force and two components of the resultant torque. These five parameters are shown to be sufficient to estimate the location of the contact point and hence the orientation of the local contact surface. Such information can then be used by the crawler’s control system for the real-time computation of an optimized foot force distribution. The intrinsic tactile sensing method has been experimentally tested on a single leg test setup, while the optimization of force distribution is already functioning in the TUM Pipe Crawling Robot (only with a different, more unripe, sensing system for the contact orientations).