Stability metrics and improved odometry prediction for tracked vehicles with tactile sensors

Abstract

In this paper, we address the motion efficiency in autonomous robot exploration with tracked vehicles in rough terrain. Tracked vehicles, along with wheel-driven propulsion systems, are the preferred platform for Unmanned Ground Vehicles (UGVs) in poor terrain conditions. However, these robots have problems with cornering, turning maneuvers or rotation around the central axis. Depending on the coefficient of friction between the tracks and the ground, the total weight and center of mass tracked vehicles produce higher slip, purely accurate and reliable pose estimation. To improve the direction of motion and the prediction of the resulting track forces and odometry calculation for tracked vehicles, a tactile surface sensor was developed to provide improved odometry determination for different ground conditions. The integration of the measurement data of the pressure sensor and the use of an improved model to determine the contact points and to improve the odometry calculation are the main objectives of this work. This is achieved by calculating the centre of gravity of the two tracks separately, using the measurement data of the pressure sensor and the local coordinates $(x,y)$ of each of the measurement points. The sensor concept was tested and evaluated on different grounds and terrains. The system can be used as a predictive model for tracked vehicle traversability and to ensure a stable position when straight manipulation tasks must be performed on rough terrain.