Vector field histogram star obstacle avoidance system for multicopters

Van Breda, Ruan (2016-12)

Thesis (MEng)--Stellenbosch University, 2016.

Thesis

ENGLISH ABSTRACT: In order to allow a multicopter to move about autonomously and detect and avoid obstacles in a two-dimensional plane, a prototype obstacle avoidance system, that implements the VFH* algorithm and uses a LIDAR sensor, was developed. The algorithm was coded in MATLAB and verified through simulations. Modifications were made to the VFH* algorithm to account for uncertainties with regard to the multicopter’s position. Thereafter, the obstacle avoidance system was successfully implemented in a test environment and managed to navigate a multicopter to its goal position for various obstacle configurations.

AFRIKAANSE OPSOMMING: Ten einde ’n hommeltuig onafhanklik te laat rondbeweeg, terwyl dit versperrings in ’n twee-dimensionele vlak waarneem en ontwyk, is ’n prototipe hindernis vermydings-sisteem, wat die VVH* algoritme en ’n LIDAR sensor gebruik, ontwikkel. Die algoritme is in MATLAB gekodeer en deur simulasies bevestig. Veranderinge is aan die VVH* algoritme aangebring om vir onsekerhede ten opsigte van die hommeltuig se posisie voorsiening te maak. Daarna is die sisteem suksesvol in ’n toetsomgewing geïmplementeer en het dit die hommeltuig suksesvol deur verskeie konfigurasies van versperrings genavigeer.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/100319
This item appears in the following collections: