Improved fruit-fly swarm algorithm for bathymetry survey by using autonomous surface vehicles /
Developing a powerful robotic system is not the only solution for solving complicated tasks. In fact, many simple swarm robots can be designed to cooperate and be able to achieve similar or even better result. It is undoubtedly more cost and time efficient to develop this simplistic swarming system....
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Kuala Lumpur :
Kulliyyah of Engineering, International Islamic University Malaysia,
2018
|
| Subjects: | |
| Online Access: | http://studentrepo.iium.edu.my/handle/123456789/4878 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Developing a powerful robotic system is not the only solution for solving complicated tasks. In fact, many simple swarm robots can be designed to cooperate and be able to achieve similar or even better result. It is undoubtedly more cost and time efficient to develop this simplistic swarming system. The system can be applied in tasks such as exploration, surveillance, and tracking. In this paper, a swarm optimization algorithm is developed to be used in autonomous surface vehicle (ASV) system in order to locate specific location within the waterbody while performing bathymetry survey. The developed algorithm is based on the existing fruit-fly optimization algorithm (FOA) and Lévy fruit-fly optimization algorithm (LFOA). These existing algorithms have been developed to be used in practical environment. However, there are several limitations that the system cannot achieve. Thus, the newly proposed algorithm is developed to overcome these constraints in order to obtain better results. The proposed algorithm is called improved-LFOA, or IFOA. It was tested and benchmarked against other several optimization algorithms such as artificial bee colony, particle swarm optimization, covariance matrix adaptation evolution strategy (CMA-ES), FOA, and LFOA. In summary, IFOA's accuracy is comparable to CMA-ES which is one of the most powerful algorithms when it comes to high-dimensional optimization. Furthermore, it also performs exceptionally well in terms of convergence rate and accuracy of the results against FOA and LFOA. In another benchmarking against LFOA in 20 restricted virtual environmental conditions, IFOA shows better convergence rate on 19 different conditions. This suggest that IFOA is a suitable algorithm for being used in ASV system for exploration task during bathymetry survey. However, the proposed algorithm still has several limitations that still need to be improved further as it still fails to operate efficiently in certain terrains. |
|---|---|
| Physical Description: | xiv, 144 leaves : colour illustrations ; 30cm. |
| Bibliography: | Includes bibliographical references (leaves 69-72). |