Wheel Terrain Interaction

Wheel terrain interaction research focuses on understanding and predicting how wheeled vehicles behave on diverse surfaces, aiming to improve robot mobility and control in challenging environments. Current efforts concentrate on developing accurate models of wheel-terrain interaction, often employing machine learning techniques like reinforcement learning and Gaussian process regression, alongside model predictive control and sliding-mode control to enhance robot performance. This research is crucial for advancing autonomous navigation in off-road and planetary exploration scenarios, improving efficiency, safety, and the reliability of robotic systems in unstructured terrains.