Optimization of autonomous driving state control of low energy consumption pure electric agricultural vehicles based on environmental friendliness

• Published in: Environmental science and pollution research international 2021-09, Vol.28 (35), p.48767-48784
• Main Authors: Zhou, Xuesheng, Zhou, Jun
• Format: Article
• Language: English
• Subjects: Accuracy; Agricultural equipment; Agricultural production; Agricultural vehicles; Agrochemicals; Algorithms; Aquatic Pollution; Atmospheric Protection/Air Quality Control/Air Pollution; Earth and Environmental Science; Ecotoxicology; Electric drives; Electric power; Energy consumption; Environment; Environmental Chemistry; Environmental Health; Environmental science; Extended Kalman filter; Herbicides; Interactive multiple model tracking; Optimization; Pesticides; Plant protection; Research Article; Spraying; State estimation; Vehicles; Waste Water Technology; Water Management; Water Pollution Control; Working conditions
• ISSN: 0944-1344; 1614-7499
• DOI: 10.1007/s11356-021-14125-9

The distributed driven agricultural vehicle platform studied in this paper is a basic carrying platform that can be used for a variety of agricultural activities (pesticide spraying, weeding, and other plant protection operations). It does not need to rely on traditional fuel engine to drive but is driven by electric power, which is more environmentally friendly and conducive to the development of ecological agriculture in the future. In this paper, according to the driving and structure characteristics of distributed driving electric agricultural vehicles, in order to ensure the vehicle trajectory tracking ability, under conditions of agricultural operation, the driving attitude and drive coordination control of electric agricultural vehicles are taken as the key research objectives. In this way, the driving path of the vehicle platform can be accurately controlled during the operation of pesticide spraying and herbicide, so as to reduce the excessive use or misuse of pesticides and herbicides and greatly reduce the field pollution. Considering the specific driving environment as well as complex and changeable motion patterns of agricultural vehicles, when a single motion model is used to track and estimate the driving state, there will be low filtering accuracy or even loss of the target during vehicle maneuvering. In this paper, interactive multiple model (IMM) algorithm is combined with extended Kalman filter to effectively track changes of the target’s motion mode, thereby avoiding low filtering accuracy or serious state estimation inaccuracy. Finally, through the distributed electric drive agricultural operation experimental platform developed by the research group, the working conditions close to the actual agricultural production activities was set up according to the needs of actual agricultural production activities in this paper, and applicability and accuracy of the algorithm state estimation are verified by experiments.