Download or read book Motion Planning for Micro Aerial Vehicles written by Sikang Liu and published by . This book was released on 2018 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: A Micro Aerial Vehicle (MAV) is capable of agile motion in 3D making it an ideal platform for developments of planning and control algorithms. For fully autonomous MAV systems, it is essential to plan motions that are both dynamically feasible and collision-free in cluttered environments. Recent work demonstrates precise control of MAVs using time-parameterized trajectories that satisfy feasibility and safety requirements. However, planning such trajectories is non-trivial, especially when considering constraints, such as optimality and completeness. For navigating in unknown environments, the capability for fast re-planning is also critical. Considering all of these requirements, motion planning for MAVs is a challenging problem. In this thesis, we examine trajectory planning algorithms for MAVs and present methodologies that solve a wide range of planning problems. We first introduce path planning and geometric control methods, which produce spatial paths that are inadequate for high speed flight, but can be used to guide trajectory optimization. We then describe optimization-based trajectory planning and demonstrate this method for solving navigation problems in complex 3D environments. When the initial state is not fixed, an optimization-based method is prone to generate sub-optimal trajectories. To address this challenge, we propose a search-based approach using motion primitives to plan resolution complete and sub-optimal trajectories. This algorithm can also be used to solve planning problems with constraints such as motion uncertainty, limited field-of-view and moving obstacles. The proposed methods can run in real time and are applicable for real-world autonomous navigation, even with limited on-board computational resources. This thesis includes a carefully analysis of the strengths and weaknesses of our planning paradigm and algorithms, and demonstration of their performance through simulation and experiments.