Phase Portrait Analysis and Drifting Control of Unmanned Tracked Vehicles

For unmanned tracked vehicles (UTVs), drifting can generate the maximum centripetal acceleration to complete challenging tasks, such as high-speed obstacle avoidance. However, due to the open-loop instability of vehicle motion during drifting, existing research on UTV motion control focuses on avoiding drifting, thereby limiting the maneuverability of the UTV. A UTV dynamics model is developed, considering the effects of acceleration, track tension and track slide on contact and shear forces. Model parameters are identified based on the moving horizon estimation method. Subsequently, the intricate nonlinear UTV model is analyzed utilizing phase portraits. Three types of equilibria for normal driving, clockwise and anticlockwise drifting are discovered, along with the region of attraction (RoA) for normal driving determined by these equilibria. An algorithm, which can simultaneously solve equilibrium curves and the RoA region (stable region for normal driving) in three-dimensional (3D) phase space, is proposed. Additionally, the saddle-node bifurcation resulting from variations in track speed differences is identified, determining the range of track speed differences. An algorithm, that can solve equilibrium surfaces (the equilibrium set) in 3D phase space, is presented. Then, a UTV drifting controller is designed to track reference states and control inputs from the solved equilibrium set, completing the high-speed continuous drifting motion. Simulation results of maneuvering a UTV model in RecurDyn and experimental results of maneuvering a scaled UTV demonstrate the fidelity of the proposed equilibrium-solving method and the effectiveness of the devised drifting controller.