From manual to autonomous: the role of intelligent technologies in the driving and control of combine harvester

Abstract

The intelligent transformation of combine harvesters is essential to the development of smart agriculture. Focusing on the transition from manual operation to autonomous harvesting, this review examines changes in the operator’s role and the technologies used to navigate and control combine harvesters. It also discusses the limitations of manual harvesting, including labor shortages, differences in operator skills, fatigue-related errors, safety risks, and inconsistent performance. Key technologies reviewed include autonomous navigation, environmental perception, mapping, crop row detection, path planning, dynamic obstacle avoidance, trajectory tracking, and artificial intelligence. These technologies reduce operator workload, improve harvesting accuracy, minimize missed areas and overlaps, and promote an operator role transition from direct driving to system supervision, emergency intervention, and fleet coordination. Major challenges include dynamic disturbances caused by changes in grain tank load, header motion, and track slippage on wet or straw-covered fields. Delays in perception, planning, and control systems can also reduce coordination among crop row detection, path replanning, header height control, and speed adjustment. In addition, existing systems do not sufficiently integrate agronomic information such as crop maturity, grain moisture, and optimal harvesting time. Data security risks in multi-machine operations may also cause collisions or incorrect grain allocation. Future advances in multi-sensor fusion, agricultural foundation models, edge computing, 6G communication, unmanned aerial vehicles, and cooperative machine groups will support autonomous and coordinated harvesting. These developments can reduce harvest losses, improve field efficiency, strengthen food security, and promote smart agriculture.