POSITION CONTROL OF OMNIDIRECTIONAL AUTOMATED GUIDED VEHICLE USING PID AND FUZZY-PID COMPENSATOR

• Main Author: Abyan Raff H, Muhammad
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/57915
• Institution: Institut Teknologi Bandung

According to World Robotics Industrial Robots 2018 report by International Federation of Robotics, demand for automated guided vehicle (AGV) has a market value of 0.5 billion dollars in 2018 and is projected to become 0.9 billion in 2021. An autonomous AGV needs a motion controller module that consists of position and velocity controller. A conventional approach to develop this module is to employ a PID-based controller. However, the ideal and linear form of PID controller may not produce satisfying result if the plant is not accurately modelled or the plant itself has a dominant nonlinear behavior. Several compensation techniques have been developed to counter this problem, one example of them is by using fuzzy controller. In the case of AGV development by PT. Irostech Solusi Intelijen, only the kinematic AGV model is available. Hence, PID-based motion controller is not enough to mitigate the error of AGV’s nonlinear movements such as curving. Therefore, in this final project an in-depth study of implementation of a fuzzy controller to compensate PID controller is conducted. The algorithm is developed inside Robot Operating System (ROS) framework in order to integrate it with the rest of AGV system, while the codes themselves are developed using python in order to ease the development phase. In this final project, the control algorithm has been developed and implemented in Robot Operating System framework. The two types of controllers have been also tested using Gazebo simulation and compared. It is recommended to implement the fuzzy compensator if the company wants more responsive behavior and faster error reduction but for pure trajectory tracking a conventional PID is enough.