Autonomous Mobile Robot Prototype for Vegetative Furrows Following Based on Optimal Control over ROS

Abstract

This study focuses on the development of autonomous reactive navigation strategies for mobile robots that must navigate between agricultural crop rows using optimal control techniques. Therefore, we configured a differential-drive mobile robot that navigates the crop row. Designed, implemented, and tested a linear quadratic regulator (LQR) and linear quadratic Gaussian (LQG) controller for a mission related to the following of crop rows. Sensors such as a laser light detection and ranging (LiDAR) sensor for distance measurement, an inertial measurement unit (IMU) that measures orientation, and encoders that measure motor angular speeds are included in the robot. To integrate sensors, actuators, and control algorithms, the robot operating system (ROS) framework was used as middleware. There is a detailed discussion of the methodology for utilizing sensors and actuators and implementing optimal control algorithms. Results obtained in this study demonstrate the effectiveness of the LQR and LQG controllers in maintaining the robot’s trajectory and correcting deviations caused by disturbances.