Visual Estimation of the Reconfigurable Mobile Robot Locomotion Effectiveness

Abstract

In this paper we suggest estimation results for the reconfigurable mobile robot locomotion effectiveness. All estimations are performed via visual analysis of the virtual modelling results. In virtual modelling scenarios mobile robot were traversing through a space with consequently assembled obstacles of five main types: slope, ledge, step, stones, and tube array. Further we describe two robotic platform configurations: tracked and wheeled. Each mobile device is assembled from a set of specialized modules. Combining of these modules allows building different mobile (wheeled, tracked, leg-wheeled and etc.) and payload systems (sensor, manipulative). All modular subsystems have virtual representations with identical to their physical analogues mass and inertial properties. We propose two different payload types: payload for wheeled vehicle (represented with modular manipulation system) and payload for tracked vehicle (represented with video camera mounted on rotation module). Furthermore we describe control system that allows virtual model of mobile robot to move between preliminary defined waypoints. As a final result we provide sequences and following descriptions of the virtual modelling animation results. Research conclusions are made as follows: based on performed visual analysis we submit that both devices are able to traverse through rough terrain with relatively small obstacles or to climb slopes with different incline angles; our further work is focused on developing more effective mobile robot configurations and expanding the class of traversable obstacles.