Design and experiment of an integrated navigation system for a paddy field scouting robot

Abstract

The navigation system is a crucial part of a paddy field scouting robot, which directly affects the working performance of the robot. The navigation error of the robot was enlarged by the inclination of the robot, which was caused by the uneven surface of the paddy fields. In this study, a map-based navigation system was proposed. The robot navigation map was derived from the planting map of a transplanter. An IMU was integrated with the GNSS to correct the navigation error using the inclination data of the robot through the Kalman filtering technique. A series of simulations and tests were performed to evaluate the performance of the integrated navigation system, including a line following test, a rough road simulation and test, and a row tracking test, as well as a field validation test. The lateral position offset and bearing angle deviation were measured and analyzed to assess the performance. The test results showed that the integrated navigation system significantly reduced the lateral position offset and bearing angle variation by 38.3 % and 26.7 % respectively as compared to the control group without the IMU correction. In the field validation test, the average lateral position offset was 5.93 cm, which was deemed acceptable for paddy field scouting operation in terms of navigation accuracy. In conclusion, the proposed integrated navigation system could be readily applied to the paddy field scouting robot to minimize navigation errors due to uneven ground and random inclination.