Advancements for A* and RRT in 3D path planning of UAVs
C. Zammit (TU Delft - Control & Simulation)
E. van Kampen (TU Delft - Control & Simulation)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Advancements in Unmanned Aerial Vehicles (UAVs) design, actuator and sensory systems and control are making such devices financially available to a wide spectrum of users with various demands and expectations. To mitigate with this ever increasing demand robust, efficient and application–specific path planning is important. This paper presents advancements over the A* and the smoothing algorithms presented in, 1 utilising the same test scenarios. Analysis of results in 1 showed a ripple in path length as the resolution changes for all scenarios considered and less than 0.1% path length improvements after certain amount of smoothing iterates. To attenuate the path length ripple, the A* ripple reduction algorithm was developed. Results show a reduction of more than 46% in terms of standard deviation with respect to the original A* algorithm without any increase in the mean path length for all scenarios. Secondly, the smoothing algorithm developed in 1 was improved to stop smoothing based on the rate of smoothing of previous iterates. Results show more than 10 multiple less path smoothing time maintaining a path length reduction especially for simple scenarios. These advancements further portray the discussed path planning algorithms as candidates to the realisation of online 3D UAV path planning.