Dynamic Collision-Inclusive Modeling of a Multirotor Aerial Vehicle using Linear Complementarity Systems

Fast and discontinuous nature of collision has raised many challenges in modeling and control of mobile robots, especially for those with agile dynamics such as multi-rotor Micro Aerial Vehicles (MAVs). In this paper, a Linear Complementarity System (LCS) model is developed to model collisions between a multirotor MAV and the surrounding environment. The contact properties and model parameters are obtained by minimizing the error between the simulation and the experimental data. The derived collision model is then verified through a series of collision experiments using a custom-designed setup. Simulation results show that the developed model is capable of estimating the post-collision velocity with the average error lower than 0.1 m/s (for maximum collision velocity of 0.91 m/s and rebound velocity of 0.4 m/s) and the maximum normal contact force errors as low as 0.25 N (≤1% of the maximum normal force).