作物表型机器人研究现状与展望

Current Status and Prospects of Crop Phenotyping Robots

摘要:

随着生物技术迅猛发展，作物育种科研对表型数据的需求日益增长，数据驱动的智能育种正成为育种研究的重要方向。高通量表型检测技术装备能够高效获取作物全生命周期表型数据，已成为制约作物规模化高效育种研究的瓶颈。作物表型机器人凭借移动灵活、作业不受时空限制，扩展性强、可挂载多种类传感器，近地多视角采集数据分辨率高，以及无人或少人操作、智能化程度高等诸多优势，是未来作物表型检测的关键发展方向。本文首先系统总结国内外作物表型机器人研究现状，阐述表型机器人整体架构，梳理其系统控制及主要导航方法，并深入介绍基于机器人的表型性状获取与解析方法，最后讨论了表型机器人在农业生产和作物育种中的应用现状及面临的挑战，指出表型机器人未来发展趋势为：机器人多样性创新将推动高通量表型检测向规模化发展，人工智能技术将重构表型解析的深度学习方法体系，而新一代表型机器人将依托多模态传感器融合技术，引领表型组学研究范式的突破。

Abstract:

With the rapid development of biotechnology, the demand for phenotypic traits in crop breeding research is on the rise, and data-driven intelligent breeding is gradually emerging as a significant direction in breeding studies. High-throughput phenotyping equipment can efficiently acquire phenotypic traits throughout the entire life cycle of crops. However, it had become a key bottleneck that restricted large-scale and efficient crop breeding research. As an emerging type of agricultural robot, crop phenotyping robots became a vital direction for future crop phenotyping due to their multiple advantages. These advantages included flexible mobility, time and space-unrestricted operation, strong expandability with the capability to carry various types of sensors, high-resolution data collection from multiple perspectives close to the ground, and high degree of intelligence enabling unmanned or minimally manned operation. Currently, there were reviews on crop phenotyping technology and the development of agricultural robots, but there were relatively few reviews specifically focused on crop phenotyping robots. The current research status of crop phenotyping robots both domestically and internationally was firstly and systematically summarized. Based on this, it elaborated on the overall architecture of phenotyping robots, sorted out their system control and navigation methods, and introduced in detail the methods of obtaining and analyzing phenotypic traits based on robots. Finally, it discussed the current application status and challenges faced by phenotyping robots in agricultural production and crop breeding, and looked ahead to the future development trend of phenotyping robots.Finally, the paper discusses the current applications and challenges of phenotyping robots in agricultural production and crop breeding, while outlining future trends characterized by three key developments: Robotic diversity innovation will propel high-throughput phenotyping toward scaled implementation, artificial intelligence will reconstruct deep learning frameworks for phenotypic analysis, and next-generation phenotyping robots leveraging multimodal sensor fusion technology will spearhead paradigm shifts in phenomics research.