In recent years, power assist suit （PAS） has been introduced in the agricultural field； however, a method to directly measure the assistive force in motion to evaluate its performance is lacking. In addition, it is difficult to say that the method of guessing it from the motor’s power change and the rotation axis can evaluate the entire PAS. Therefore, in this study, we developed a method to dynamically measure the assistive force with load on PAS. Consequently, we clearly expressed the change in assistive force to the rotation angle ; after five measurements, the standard deviation of the same rotation angle gained high repeatability at an average of 0.2 N m.

Operating the electric wind winnowing unit of a combine using continuously variable transmission allows fine adjustment of the wind winnow, resulting in improved versatility and quality of winnowing. Additionally, this may aid in devising solutions for problems resulting from the automation of combines. In this study, the changing point in the grain flow rate was determined based on the obtained load currents before and after the grain flow rate was changed via state-space modeling. Moreover, the performance of the wind winnow, in terms of its sorting accuracy, was classified as good or bad using discriminant analysis, which was further validated using cross-validation. Based on the results, a detection algorithm was developed.

An estimation of the percentage of ripened grains requires measuring the volume and the mass of a rough rice layer in a combine grain tank. Helmholtz resonance can be used to measure the volume. However, its measurement accuracy lowers because the sound pressure attenuates through pores in a rough rice layer. This study clarified the characteristics of viscous damping coefficients α used in the volume estimation equation considering the attenuation and evaluated the volume estimation accuracy. Coefficient α, which had a strong liner relationship with the pore volume (R² ≥ 0.975), was estimated using rough rice mass (R²＝0.988). The relative error of volume estimates was generally within ±5 %, greatly improving from −20 % using the conventional equation without considering the attenuation.

In this study, transfer path analysis (TPA) technique was employed to estimate the best reference points to the response point of a hand tractor. Six locations of the hand tractor, specifically, engine-right-side front, engine-right-side rear, engine-cover top, chassis, gearbox, and middle between handle-base and handgrip were considered as reference points; whereas, handgrip was regarded as a response point. For TPA analysis, the power spectrum densities of the seven locations were populated into a 3rd-order tensor. To improve the accuracy of response point vibration estimation, two reference points were selected for TPA analysis. Finally, the best contribution was determined using the root mean square error (RMSE) methods. Result indicated that chassis-gearbox combination are the best target points for the response point.