Learning the Forward Predictive Model for an Off-Road Skid-Steer Vehicle

A forward predictive model is used to simulate a vehicle's motion given a sequence of commands that could potentially be executed. Generally, forward predictive models are used by planning systems for Unmanned Ground Vehicles (UGV’s) so that commands can be selected such that obstacles are avoided. This report presents a datadriven approach for learning a forward predictive model based on previously recorded vehicle motion. The selected approach is compared to several variations including the conventional forward predictive model that has traditionally been used on the Crusher vehicle. Results are presented using real life data collected on the Crusher UGV