Model predictive control-based trajectory planning for quadrotors with state and input constraints
Penghong Lin, Songlin Chen, Chang Liu · 2016
A novel scheme to solve the trajectory planning problem for quadrotors with model and state constraints is proposed. First the RRT (Rapidly-exploring Random Tree) algorithm is employed to generate an initial route in context of a 3D environment. Then with the model of quadrotoer, the MPC (Model Predictive Control) method is used to construct an inner simulator which can generate the trajectories satisfying the state constraints by following the waypoints exported by the RRT solution. Finally, the trajectory generation of quadrotors with state and input constraints is transformed into a general form of QP problem, which can be solved expediently by common numerical algorithms. The proposed scheme can generates high-quality and feasible trajectories satisfying both state constraints and input constraints, which is verified sufficiently by extensive simulations.