Extended bat algorithm for PID controller tuning of wheeled mobile robot and swarm robotics target searching strategy

• Main Author: Nur Aisyah Syafinaz, Suarin
• Format: Thesis
• Language: English
• Published: 2020
• Subjects: TJ Mechanical engineering and machinery; TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://umpir.ump.edu.my/id/eprint/30376/1/Extended%20bat%20algorithm%20for%20pid%20controller%20tuning%20of%20wheeled%20mobile%20robot.pdf; http://umpir.ump.edu.my/id/eprint/30376/

Swarm robotics is a multi-robot system which consists of more than two robots working together to accomplish a specific task. Swarm robotics has a huge potentiality to be applied in many sectors such as in agriculture and military due to the availability number of manpower in the system and the capability of robots to execute and accomplish task faster than manual labor. Swarm robotics system consists of at least two robots and can up to hundreds and thousands number of robots in the swarm. Thus, there is an urgent need to monitor and manage the movement of all robots in order to avoid any collision and accident among robots and at the same time to optimize the performance of swarm robotics system. In order to build a swarm robotics system that able to perform target searching task, the system needs to have a robust strategy. However, current solutions of target searching strategy for swarm robotics are unsatisfactory as the solutions are unable to make the robots move efficiently and not arrive at the target position precisely. There are many factors that might contribute to that condition e.g. controller, algorithm and communication. Each robot in the swarm needs to be equipped with good controller, e.g. robust PID controller. Although PID controller is a traditional controller with simple architecture, tuning PID controller to the best gains in order to develop good performance system is difficult. Thus, in order to identify and provide the solution, this research study decided to start the approach with the optimization of PID controller of Wheeled Mobile Robot (WMR). This is because, WMR must able to move effectively from one point to another point in order to execute the target searching task and to perform well in swarm robotics system. The objectives of this research study is 1)to tune and optimize gains of Proportional-Integral-Derivative (PID) controller for wheeled mobile robot (WMR), 2)develop target searching strategy for swarm robotics system and 3)compared the performance of proposed method with the well-established methods, target searching strategy based on Particle Swarm Optimization (PSO) and Bat Algorithm (BA). Extended Bat Algorithm (EBA) has been chosen as swarm intelligent based method for this research study. EBA is the hybrid swarm optimization method of BA with SDA. The weakness of BA to explore the solution appropriately has been overcome with SDA technique. The method of the research study is initiate with tune PID based controller for WMR. Next, adaptive EBA is applied to develop target searching of swarm robotics for two cases i.e. 1) different initial position and varies distance of robots in the swarm and 2) different population of robots in the swarm. Last but not least, the performance of proposed algorithm is compared with PSO and BA. Results obtained from this research study indicates that PD controller tuned by EBA outperformed the performance of PID and PI controller by of 11.00s and 12.11s of rise time for X and Y position respectively, 20.08s and 22.08s settling time X and Y position respectively and no overshoot for both position. By applying the developed robot with PD controller, EBA prove its potentiality to develop the best target searching strategy to the swarm robotics system with 5 number of iterations within 49 seconds. The result is the lowest number of iterations in the shortest of time. The accuracy is 99% to arrive at the desired location compared with target searching strategy based on PSO and BA. Hence, EBA which is the proposed method is able to develop target searching algorithm for swarm robotic system and tuning PD controller gains for wheeled mobile robot.