To sustain food production, the introduction of robotics and automation in agricultural operations has become increasingly important. This study aimed to automate tractor tillage operations. We developed a system that automatically controls tractor speed by mounting a soil pulverization ratio sensor on the implement (rotary tiller) and sending control commands from an onboard computer to the tractor based on real-time measurements. Field experiments were conducted to evaluate system performance. Compared with conventional tillage, the proposed system increased the proportion of observations in which the soil pulverization ratio was within the predefined target range from 59% to 74% and reduced the variance of the soil pulverization ratio from 26.5 % 2 to 18.6 % 2 . The control loop was executed at 10 Hz during field operation. In practical field operations, switching among different tractors and implements is common; therefore, the system should operate across different tractor models. To achieve this, the system was extended to support two communication protocols, including Tractor Implement Management (TIM), an ISOBUS-based function, and EcoTraLink. Experiments on speed control were conducted using different tractor models to evaluate transient characteristics. The results revealed substantial differences in dynamic response between tractors. This finding highlighted a limitation of current tractor–implement communication protocols: although command messages are defined, neither the response characteristics nor a mathematical model of tractor dynamics are specified. This limitation makes it difficult to apply advanced model-based control while maintaining interoperability. Finally, potential approaches to addressing this limitation were proposed.
Tsuchikawa et al. (Mon,) studied this question.