Active Informative Receding Horizon Planning for Quadrotors

Motion planning for quadrotors is still a challenging problem due to the uncertainty and complexity of the environment. In this context, we propose an active informative receding horizon trajectory generation method that incorporates the informative path planning and B-spline-based trajectory optimization to achieve a smooth, safe, and efficient trajectory for UAVs. The proposed method includes an Informative A* (IA*) path planning algorithm, which not only considers the path length but also takes into account the perception information. Specifically, the informative term is calculated by transforming the information matrix of the least-squares problem based on the odometry, which can reflect the environmental texture. The trajectory is then generated based on a B-spline curve, which takes into account collision avoidance, smoothness, and dynamic feasibility. Finally, comparative simulations are conducted to verify the superior performance in improving the localization accuracy.