Target Detection in UAV Automatic Landing Based on Artificial Bee Colony Algorithm with Bio-Inspired Strategy Approach

• Published in: Robotics and Rehabilitation Intelligence pp. 253 - 271
• Main Authors: Zhang, Beiwei, Cao, Jiangtao, Guo, Ying
• Format: Book Chapter
• Language: English
• Published: Singapore Springer Singapore
• Series: Communications in Computer and Information Science
• Subjects: Artificial Bee Colony algorithm; Chaotic search; Swarm intelligence; Unmanned Aerial Vehicles
• ISBN: 9789813349285; 981334928X
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-981-33-4929-2_18

To meet the requirements of detecting runway and salient ground targets precisely and rapidly for Unmanned Aerial Vehicles (UAV) during automatic landing, a scenario that integrates our modified Artificial Bee Colony (ABC) algorithm with template matching is proposed in this paper. This study aims at enhancing the traversal search capability and avoid evolutionary stagnation with the Bio-inspired ABC (BABC) algorithm. Firstly, the initial solution optimization strategy of reverse learning is adopted in the initialization phase. Secondly, an improved search strategy with Levy flight characteristics is applied in the phase of updating the population and local searching. Additionally, chaotic searching strategy is introduced into ABC to reinforce the ability of system and get rid of local optimal solution, thus showing advantage of convergence property and robustness when compared with ABC. Series of experimental results demonstrate the feasibility and effectiveness of our presented approach over the standard method. The novel algorithm also meets the requirements of speed and precision in practical application.