An Improved Model Predictive Trajectory Planning Method for Unmanned Tracked Vehicles on Undulating Terrains

To ensure the safety and stability of unmanned tracked vehicles (UTV) driving under off-road scenarios, an improved model predictive trajectory planning method for UTV navigating on undulating terrains is proposed. Firstly, a terrain-based kinematic model is established to consider the impact of terrain features on the UTV attitude. Secondly, based on the model mentioned above, an improved model predictive trajectory planning (MPTP) method is proposed. This method can solve a trajectory that satisfies the motion constraints under undulating terrain scenarios. Thirdly, to reduce the computational complexity of the rolling optimization, an initial guess for MPTP is generated via a modified hybrid A* method, in which terrain features are considered. The proposed method has been validated in a typical undulating terrain scenario, and the results indicate that it performs better than conventional trajectory planning method for UTVs under the undulating terrain scenario.