Research progress and development trend of bionic harvesting technology

• Published in: Computers and electronics in agriculture Vol. 222; p. 109013
• Main Authors: Luo, Yuanqiang, Li, Junlin, Yao, Beihuo, Luo, Qing, Zhu, Zhicheng, Wu, Weibin
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.07.2024
• Subjects: Agricultural biomimicry; agriculture; algorithms; automation; Bio-inspired algorithms; Bionic harvesting; electronics; End-effector; natural selection; neural networks; system optimization; Bio-inspired algorithms; Bionic harvesting; End-effector; Agricultural biomimicry
• ISSN: 0168-1699; 1872-7107
• DOI: 10.1016/j.compag.2024.109013

•This review fills a gap in the application of bionics to agricultural harvesting.•The applications of bionic algorithms in agricultural harvesting were summarized.•The key experimental means of bionic harvesting technology were summarized.•Challenges and future trends of bionic harvesting technology are reported. In the context of natural selection, organisms have evolved various excellent characteristics that can guide improved design optimization in engineering. Scholars at home and abroad have carried out a large number of efficient and favorable bionic-type researches by imitating the information of biological signs. In particular, in the field of agricultural picking, the bionic concept of “learning from nature and feeding back to nature” will undoubtedly facilitate the automation of various picking tasks in agriculture, which can not only solve the social problems and operational efficiency problems caused by manual harvesting, but also effectively reduce the damage rate of target crops. Based on these, the present study conducted a systematic literature review of the latest advancements in bionic technology implementation for agricultural picking, utilizing multiple scientific databases and online development resources. In particular, the bionic end-effector structures and bionic-inspired algorithms (BIAs) are deeply studied, and both types of bionic applications are well categorized. The bionic end-effector structures are considered in this study to be divided into three main categories: (1) cutting type, (2) clawing type, and (3) clamping type. The BIAs mainly consider four common categories: (1) artificial neural network, (2) genetic algorithm, (3) artificial bee colony, (4) ant colony optimization. In addition, this review also explores the experimental treatments used to develop the bionic picking structure, which aims to guide how to learn the excellent characteristics of biological signs to optimize the mechanical design. Finally, the potential challenges and future development trends of bionic harvesting technology are prospected.