To enhance the path tracking accuracy and dynamic adaptability of small unmanned harvesters in complex farmland environments, this paper proposes a simulation and autonomous driving algorithm framework based on TruckSim and Simulink. By innovatively integrating TruckSim’s high-precision dynamic simulation with Simulink’s powerful algorithm development capabilities, we have constructed a comprehensive simulation platform that accurately models the harvester’s behavior in agricultural settings. This platform not only accurately simulates dynamic responses under various operating conditions but also facilitates efficient testing and validation of autonomous driving algorithms, thereby significantly shortening development cycles and lowering field-testing costs. For path planning, we implement a hybrid A* algorithm with dual heuristic search strategy to generate optimal paths in typical static agricultural operations. At the control level, a PID controller is designed to optimize path tracking and speed control performance. Furthermore, an Extended Kalman Filter-based road adhesion coefficient identification method is introduced, which integrates multi-sensor data to dynamically estimate road conditions and adjust control strategies accordingly. To enhance system robustness, a PID-based lane-keeping algorithm with steering-speed coordination mechanism is incorporated, significantly improving operational stability in various farmland environments. Field validation results demonstrate that this research provides an innovative simulation tool and effective algorithm validation platform, advancing the development of intelligent agricultural equipment.
Sun et al. (2026) studied this question.