Development of angular correction algorithm for movement of agricultural mobile robots in a straight line

Abstract

This paper proposes the angular correction algorithm for the autonomous navigation of the agricultural mobile robots, which are driven in a straight line, with simple hardware based on the data of the digital compass and the GPS receiver. The motion of the mobile robot was accomplished by the differential drive mechanism with four driving wheels in which the overall velocity is split between left and right wheels. The two-channel DC motor controller was used to drive motors. The digital compass was used to calculate the position angle of the mobile robot with respect to the target point. A Kalman filter was used to fuse the information from GPS and digital compass. In the proposed algorithm, the mobile robot is driven in a straight line along a predefined path by calculating in real time the deviation angle difference with respect to the target point. When the robot encounters an unexpected external force varying the desired path, it achieves a smooth and stabilized straight line movement by correcting the deviation angle difference. The performance of the mobile robot was evaluated based on a total of 18 straight lines in a farmland. Standard errors of cross track error (XTE) values of straight lines for each target point were analyzed. The mean of arithmetic means was found to be 4.14 cm. The mean of Rsquare was 0.990. This value shows that the proposed angular correction algorithm is useful in driving the mobile robot in a straight line.