Gait identification and optimisation for amphi-underwater robot by using Ant Colony Algorithm /
Manoeuvrable robot commonly has become the focus of the latest heated issues especially in applications that involved disaster rescue, military missions and underwater or extra-terrestrial explorations. Currently, the manoeuvrable robot is controlled manually by the operator and it's a wheeled...
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/123456789/4834 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Manoeuvrable robot commonly has become the focus of the latest heated issues especially in applications that involved disaster rescue, military missions and underwater or extra-terrestrial explorations. Currently, the manoeuvrable robot is controlled manually by the operator and it's a wheeled type. It is used for rescue missions to transport people from disaster area to the safe zone. However, the robot is incapable of moving automatically, and it goes through terrain or landscape like swarm. Therefore, a suitable platform is required to transport or for other uses especially in dangerous mission. It is very difficult to estimate the movement of the robot to avoid obstacles and choose the alternative path. Hence, this research presents the point-to-point gait identification of the behavious of the robot to manuever autonomously on both on-land and underwater environment. For the optimization, the robot will travel from one specific point to another with the predefined position within optimized gait and fastest time by using Ant Colony Optimization (ACO) technique. The algorithm being compared, between Ant Colony Algorithm (ACO) and the Particle Swarm Optimisation (PSO) in terms of time and distance. The ACO been chosen because of the positive feedback for rapid discovery and able to use in dynamic applications for example adapts to changes like new distances. The performance of the algorithm showed that the execution time of ACO is more realistic. Hence, Matlab software is used to determine the best cost extracted from the ACO with the pre-define of number of iteration and the number of ants. It also become the controller with the integration between Matlab data with the Arduino as a controller for prototype. with different types of input, such as number of ants, points and iterations. The results showed that with the higher number of points give the best cost however it increases the time execution. The laboratory-scaled prototype was developed to test the design controlled with ACO technique where Global Positioning System (GPS) is used for the coordination of the robot and Magnetometer for the position of the robot. The prototype was first modelled by using virtual reality in Solidworks Design. The prototype is builded for amphi-underwater purpose which is able to move on both land and water. The purpose of this robot is able to overcome obstacle such as swarm, stone and others and able to move autonomously with the predefined point. The simulation performance analysis has been conducted to evaluate the robustness of the designed with the predefined area. As a final result, this autonomous robot should be able to maneuver autonomously and can be switched to manual mode during the experiments as well as able to overcome all the obstacle under the certain conditions. However, to become autonomous, it needs a stable algorithm that can be configured and can be implemented in the robot. In order to do so, the algorithm has been predefined to the initial condition and final destination which faced problems to overcome the obstacles and is compared with the real robot. Based on the results, it shows that the best ten time of trials autonomous robot only has difference for 26.32 seconds with the distance of 20 meters. The robot prototype is able to move autonomously and optimized by the ant colony optimization with predefined position and terrain condition. |
|---|---|
| Physical Description: | xix, 99 leaves : colour illustrations ; 30cm. |
| Bibliography: | Includes bibliographical references (leaves 83-87). |