Control System for Mobile Robot using FPGA-Based Q-Learning Accelerator

🔗

Abstract: Q-Learning algorithm is used in smart navigation, path planning and others. The output of the Q-Learning cannot be directly used to drive motors or control robot to specific function in the testing area. To convert Q-Learning action into low level signal, we need a controller that organize the task of the sensors, the actuators and the main accelerator. In this paper, we present the control system for a mobile robot that uses FPGA-Based Q-Learning Accelerator. The control system is connected to the sensors, actuators and the Q-Learning Accelerators. The control system is successfully connected to the FPGA development board and deployed to the mobile robot. Moreover, the simulation result of the control for the actuator shows stability. The control system is useful for designing application systems such as smart navigation, IoT, and smart communication.

I. Syafalni, M. I. Firdaus, N. Sutisna, Y. W. Hadi and T. Adiono, “Control System for Mobile Robot using FPGA-Based Q-Learning Accelerator,” 2022 8th International Conference on Wireless and Telematics (ICWT), Yogyakarta, Indonesia, 2022, pp. 1-6, doi: 10.1109/ICWT55831.2022.9935386.

URL: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9935386&isnumber=9935131