Intelligent Terrain-Aware Navigation and Mobility of Unmanned Ground Vehicles Operating Under Varying Degrees of Autonomy

Robotic Research, LLC in partnership with Tulga Ersal at the University of Michigan have joined forces under Army STTR topic A14A-T018 Intelligent Terrain-Award Navigation and Mobility of Unmanned Ground Vehicles Operating Under Varying Degrees of Autonomy to develop a remote teleoperation* system for driving unmanned vehicles off-road and at high speed. This system shall remain effective in the presence of delays in communications between the operator and the vehicle. To ensure effectiveness, we start with existing state-of-the-art mobility and path planning technology and develop a terrain-aware operator-vehicle system with a high degree of autonomy. Terrain awareness and autonomy will make remote teleoperation of unmanned assets safer and faster by taking steps to minimize risk from dynamic vehicle-terrain interactions (VTI). The novelty of our approach lies in integration of detailed online terramechanics models on both ends of the system (operator and vehicle) and using a new terrain estimation method.We plan to evaluate the use of these models as a means to ensure that semi-autonomous driving at high speed is safe, meaning the vehicle will not find itself in a hazardous situation where it would risk rolling over, getting stuck, spinning out of control or crashing into an obstacle.