An automatic forage unloading method based on machine vision and material accumulation model

The traditional cooperative operation method of forage harvesters and trailers suffers from low hopper unloading rates and forage waste, which hinders the development of precision agriculture. To address these issues, this study proposed an automatic forage unloading method. First, the outline of the forage pile was extracted based on machine vision, and the pile's repose angle was determined by linearly fitting its profile. Next, we utilized the forage accumulation model to develop a prediction method for the forage drop point and the automatic unloading method. The reliability of the forage accumulation models and the feasibility of the automatic unloading method were then verified through DEM simulation. Finally, field forage harvests were conducted to validate the effectiveness of the automatic unloading method and the accuracy of the mass calculation model. The simulation showed that the average absolute errors of the simulation and model calculation results of the three material piles were less than 0.1 m3, respectively. The field experiment showed that the maximum absolute error between the predicted and actual drop points was 0.095 m. The average absolute errors of the mass calculation model and the weighing method were 21.37 kg, 37.68 kg, and 60.36 kg, respectively. This research improves harvesting efficiency, reduces forage waste, and promotes the automatic development of self-propelled forage harvesters.