Quick turn headland-turning control strategy for autonomous rice transplanters

Abstract

Efficient headland turning is crucial for maximizing the productivity and accuracy of autonomous rice transplanters, especially in small-scale fields where frequent maneuvering is necessary. This study developed and experimentally validated a practical headland-turning strategy that enables rapid, continuous turns without requiring reverse maneuvers or vehicle stops. Two complementary techniques were employed: a dynamic waypoint adjustment using real-time circular regression and an experimentally validated lateral boundary offset (LBO) model. These were designed to reduce trajectory deviations during transitions between turning and straight-line motion. Field tests with a commercial autonomous rice transplanter demonstrated successful turns at various speeds, emphasizing the need to account for field-specific conditions such as surface irregularities caused by prior tillage. The results confirm the effectiveness of the proposed strategy in improving precision and efficiency, providing a strong basis for future advancements in autonomous transplanting systems.