Research Progress of Apple Production Intelligent Chassis and Weeding and Harvesting Equipment Technology
DUAN Luojia,YANG Fuzeng,YAN Bin,SHI shuaiqi,QIN jifeng
DOI: https://doi.org/10.12133/j.smartag.SA202206010
2022-01-01
Abstract:As a pillar industry of economic development in the main apple-producing areas, apple industry has made important contributions to the increase of local farmers' income. With the transformation and upgrading of apple industry, the mechanization and intelligence level would be directly related to economic benefits. To promote the research of apple production intelligent technology and the development of intelligent equipment, in this paper, the current level of mechanization in each step of apple production was first introduced. Then, the main characteristics of the main apple orchard machinery, such as power chassis, weeding machinery, and harvesting equipment, were demonstrated. The application progress of automatic leveling and control, automatic navigation, automatic obstacle avoidance, weed identification, weed removal, apple identification, apple positioning, apple separation, and other technologies in intelligent power chassis, intelligent weeding machines, and apple harvesting robots, were summarized. The basic principles and characteristics of the above three key technologies of intelligent equipment were expounded in combination with different application environments. Intelligent control is the key technology for the intelligentization of orchard power chassis. The post of chassis adaptive control technology and autonomous navigation technology were discussed. In addition, a chassis intelligent perception and intelligent decision-making system should be established. Orchard chassis safe, accurate, efficient, and stable driving and operation is the future development trend of orchard intelligent chassis. The lack of robust weed sensing technology is the main limitation to the commercial development of a robotic weed control system. To improve the level of weed detection and weeding, machine vision and multi-sensor fusion methods have been proposed to solve the practical problems, such as illumination, overlapping leaves, occlusion, and classifier or network structure optimization. Robotic apple harvesting has proven to be a highly challenging task due to environmental complexities, sensor reliability, and robot stability. To improve the accuracy and efficiency of harvest mechanization applications in apples, apple quick identification under complex scenes, apple picking path planning, and materials and structure of manipulator for apple picking must all be optimized accordingly. Finally, the challenges of intelligent equipment technologies in apple production were analyzed, and the developing suggestions were put forward. This research can provide references and ideas for the advancement of intelligent technology research in apple production and the research and development of intelligent equipment.