Kinodynamic planning for a fixed-wing aircraft in dynamic, cluttered environments : a local planning method using implicitly-defined motion primitives

[摘要] ENGLISH ABSTRACT: In order to navigate dynamic, cluttered environments safely, fully autonomous UnmannedAerial Vehicles (UAVs) are required to plan conflict-free trajectories between two statesin position-time space efficiently and reliably. Kinodynamic planning for vehicles withnon-holonomic dynamic constraints is an NP-hard problem which is usually addressedusing sampling-based, probabilistically complete motion planning algorithms. These algorithmsare often applied in conjunction with a finite set of simple geometric motionprimitives which encapsulate the dynamic constraints of the vehicle. This ensures thatcomposite trajectories generated by the planning algorithm adhere to the vehicle dynamics.For many vehicles, accurate tracking of position-based trajectories is a non-trivialproblem which demands complicated control techniques with high energy requirements.In an effort to reduce control complexity and thus also energy consumption, a genericLocal Planning Method (LPM), able to plan trajectories based on implicitly-defined motionprimitives, is developed in this project. This allows the planning algorithm to constructtrajectories which are based on simulated results of vehicle motion under thecontrol of a rudimentary auto-pilot, as opposed to a more complicated position-trackingsystem. The LPM abstracts motion primitives in such a way that it may theoretically bemade applicable to various vehicles and control systems through simple substitution ofthe motion primitive set.The LPM, which is based on a variation of the Levenberg-Marquardt Algorithm (LMA),is integrated into a well-known Probabilistic Roadmap (PRM) kinodynamic planning algorithmwhich is known to work well in dynamic and cluttered environments. The completemotion planning algorithm is tested thoroughly in various simulated environments,using a vehicle model and controllers which have been previously verified against a realUAV during practical flight tests.

[发布机构] Stellenbosch University