Design and Optimization of Motion Control for Autonomous Scrubber Robot

Abstract

This work describes the preliminary design and development of an autonomous scrubber robot with the compatibility of the robot operating system (ROS). In the development of autonomous robots, the critical challenge lies in devising an advanced motion control system that can mitigate inherent uncertainties, such as wheel asymmetry, sensor inaccuracies, and external disturbances, to ensure precise and consistent robot movement, especially in tasks requiring straight-line trajectories. The utilization of the PID algorithm enables the accurate control of the speed of two brushless DC motors in the autonomous scrubber robot. The system architecture incorporates an Arduino Mega 2560 microcontroller board, along with the integration of a two-axis gyroscope and accelerometer for determining the system's orientation. The integration of Advanced Monte Carlo Localization (AMCL) within the ROS framework enhances the system's ability to accurately determine the robot's pose by leveraging probabilistic methods and real-time sensor data integration. During the experiments, a noticeable trend emerged in the observed error percentages as the robot's covered distance increased. Experimental results demonstrate the machine's ability to maneuver flexibly, including straight-line movements.