OPTIMAL CONTROLLER DESIGN FOR LEADER-FOLLOWER SYSTEM BETWEEN UGV AND UAV

Abstract

Unmanned vehicles are already in wide use. For instance, they can be applied for military purposes, space exploration, in routine life, delivery services, etc. Unmanned vehicles are in great demand since they can be exploited at places that are dangerous or unachievable for humans. This research aims to create a leader-follower system for unmanned aerial vehicle (UAV) and unmanned ground vehicle (UGV) that could have many applications. The drone will fly ahead and scan the surface from above for information. While the robot will create its own route in order to get to the point determined by the drone. This can be used to explore hard-to-reach places where the drone is conducting reconnaissance, and the robot is collecting materials with a robotic arm. For the project, the Quanser system was used, which includes a tracking system, communications, QDrone as a UAV, and QBot 2e as a UGV. A fuzzy controller was created for QDrone, which ensures flight stability. This thesis details the dynamic models of vehicles and the subtleties of tuning the controller. In addition, methods for creating a leader-follower algorithm are described in detail. In addition to everything, an obstacle avoidance algorithm for QBot 2e was created, which was added to the leader-follower algorithm. The experiments conducted in this thesis with Fuzzy, PID and Fuzzy with prefilter controllers have shown that the Fuzzy controller performs more accurately and better in controlling the position of the quadcopter. An algorithm for the Leader-Follower system with obstacle avoidance was developed, and experiments were conducted to compare the Fuzzy and PID controllers under different conditions. Finally, the optimal Fuzzy controller was integrated into the Leader-Follower system with a Fuzzy controller. The results show that the proposed controller is effective in controlling the system and avoiding obstacles.