农业收获装备自动对行关键技术研究进展

Research Progress on Key Technologies of Automatic Row Guidance for Agricultural Harvesting Equipment

摘要:

自动对行是提升农业收获装备作业精度、降低机收损失和驾驶员劳动强度的重要技术。针对复杂田间环境下作物行感知稳定性不足以及控制精度与实时性难以兼顾等问题,本文分析了该领域关键技术与研究进展,围绕“感知—决策—执行”闭环控制架构,系统梳理了机械触觉、机器视觉、激光雷达及多源信息融合技术的感知原理、适用场景与典型案例;归纳了从PID控制到模型预测控制、模糊控制及深度强化学习等导航控制算法的演进逻辑;分析了电液比例转向、伺服电动纠偏及工作部件横移等执行机构的响应特性。结合自动对行技术典型应用场景与案例,指出了自动对行技术在农业收获装备应用中还存在的问题,并展望了该技术未来的主要发展方向为:推进多源传感器互补融合与轻量化实时识别以提升复杂工程感知适应性,发展状态自适应、安全容错的自主决策控制方法,强化感知、决策、控制与执行的协同匹配及可靠性验证,并通过低成本模块化配置、标准化接口和多区域长周期评价促进技术普惠应用。

Abstract:

Automatic row guidance is an important technology for improving the operational accuracy of agricultural harvesting equipment, reducing mechanical harvesting losses, and alleviating operators’ workload. To address the insufficient stability of crop-row perception under complex field conditions and the difficulty in balancing control accuracy and real-time performance, the key technologies and research advances in this field were analyzed within a closed-loop control architecture consisting of perception, decision-making, and actuation. The sensing principles, applicable scenarios, and representative applications of mechanical contact sensing, machine vision, light detection and ranging (LiDAR), and multisensor information fusion were systematically reviewed. The development of navigation control algorithms, including proportional-integral-derivative ( PID) control, model predictive control, fuzzy control, and deep reinforcement learning, was summarized. The response characteristics of electro-hydraulic proportional steering, servo motor-based correction, and lateral shifting mechanisms for working components were also analyzed. Based on typical application scenarios and representative cases of automatic row-guidance technology, the remaining problems in its application to agricultural harvesting equipment were identified, and the main directions for future development were discussed.