Abstract

Overall navigation in orchards is essential for autonomous production activities, however, most navigation studies, in orchards, focus on in-row scenes, and few of them pay attention to the headland transition or the overall navigation. This paper proposes an integrated navigation method based on satellite and lidar for overall orchards scenes, where the lidar was mainly used for in-row navigation and chassis position (CP) obtaining, whilst the satellite was utilized for headland transition as well as the assistant for in-row integrated navigation. Two overall integrated navigation modes were investigated: CMD-INT, which integrates the command signals from the lidar-based and satellite-based navigation systems, and BIAS-FUS, which combines the information of bias from the two navigation systems. The CP was used to detect the headland of the tree rows and achieve overall navigation integration. A position-ratio-based path planning method was also proposed, which enabled the chassis to maintain a proper constant offset from one side of the orchard. The refined pure pursuit algorithm was adopted for path tracking with good adaptivity on both the path curvature and tracking velocity. Both simulation and field tests with ten group parameters configuration in total were conducted to verify the performance of the system. The BIAS-FUS mode outperformed the CMD-INT mode, and the findings for the simulation experiments were consistent with that of the field tests. In the simulation experiments, the maximum value (Max), mean value (Mean), and root mean square error (RMSE) for the in-row process decreased by 0.0054 m, 0.0053 m, and 0.0062 m, respectively. Additionally, the corresponding heading deviation decreased by 0.3027°, 0.0050°, and 0.0256°, respectively. In the field tests, the Mean of the heading and lateral deviation for the in-row tracking process decreased by 0.3823°, and −0.0072 m respectively, and the corresponding deviations for the transition tracking process decreased by 2.8903°, and 0.0164 m respectively. Aiming at the standardized tall spindle apple orchards, this work contributes to an available overall integrated navigation method.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call