A self-propelled machine for combined potato harvesting and residual plastic film retrieval is presented in this paper. The machine was designed collaboratively and built at the College of Mechano-Electronic Engineering of Gansu Agricultural University. It is intended for slow slope and horizontal terraces in hilly and mountainous areas of Northwest China, where regular-size harvesters cannot operate. The machine can realize the combined operations of potato digging, potato separation from soil and plastic film, potato collection and bagging, and residual plastic film retrieval. Through engineering analyses, the main systems of the machine were calculated, and their operating parameters were estimated. These include the digging and lifting device, the potato–plastic-film separation device, and the residual plastic film retrieval device. Field tests were performed at a 0.5 m/s driving speed of the machine, while the linear speed of the lifting chain of the digging and lifting device was 1.5 m/s, the tilting angle of the conveying chain of the potato and plastic film separation device was 50°, its linear speed was 0.6 m/s, and the linear speed of the lifting screen of the circulating lifting device was 0.7 m/s. With these settings, the average productivity of the machine was 0.12 ha/h. The loss rate, damage rate, and potato bruising rate were 1.8%, 1.4%, and 2.8%, respectively; the potato impurity rate was 3.6%; and the residual plastic film retrieval rate was 83%—all above industry standards. This research provides a solution to the problem of mechanized potato harvesting and plastic mulch retrieval on small, slopped plots of land in Northwest China and in other parts of the world where similar conditions exist.

The seed metering process of a fluted force-feed seeder was simulated using the Discrete Element Method and its parameters optimized using the Box–Behnken Design of Experiments and the Response Surface Method. The rotational speed of the feed roller, the lead (helix) angle of the flutes, and the number of flutes were the independent variables, while the response value was the seeding uniformity index. Two regression models were investigated, and the following conclusions drawn. For the flute lead angle between 0 and 10 degrees, and the number of flutes between 10 and 14, it was found that the number of flutes and the lead angle influenced the seeding performance the most, with the order of importance being the (i) number of flutes, (ii) lead angle and (iii) roller speed. For the flute lead angle between 5 and 15 degrees, and the number of flutes between 12 and 16, it was found that the roller speed and the number of flutes influenced the seeding performance the most, with the order of importance being the (i) roller speed, (ii) number of flutes and (iii) flute lead angle. The two regression models were then minimized for the seeding uniformity index and the corresponding optima verified experimentally on a conveyor belt test stand fitted with an image recognition system.