Q-Learnıng Based Real Tıme Path Plannıng for Mobıle Robots

Decision making and movement control are used for mobile robots to perform the given tasks. This study presents a real time application in which the robotic system estimates the shortest way from robot's current location to target point via Q learning algorithm and makes decision to go the target point on the estimated path by using movement control. Q Learning algorithm is known as a Reinforcement Learning RL algorithm. In this study, it is used as a core algorithm for estimation of the path that is optimum way for mobile robot in an environment. The environment is viewed by a camera. This study includes three phases. Firstly, the map and the locations of all objects including a mobile robot, obstacles and target point in the environment are determined by using image processing. Secondly, Q Learning algorithm is applied for the problem of the estimation of the shortest way from the current location of the robot to target point. Finally, a mobile robot with three omni wheels was developed. Experiments were carried out using this robot. Two different experiments are performed in experimental environment. The results obtained are shared at the end of the paper. Halil Cetin | Akif Durdu | M. Fatih Aslan | M. Mustafa Kelek "Q-Learnıng Based Real Tıme Path Plannıng for Mobıle Robots" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-4 | Issue-1 , December 2019, URL: https://www.ijtsrd.com/papers/ijtsrd29625.pdf Paper URL: https://www.ijtsrd.com/engineering/computer-engineering/29625/q-learn%C4%B1ng-based-real-t%C4%B1me-path-plann%C4%B1ng-for-mob%C4%B1le-robots/halil-cetin