基于FPGA动态视觉的植保无人机精准施药系统设计

Design of a Precision Spraying System for Agricultural Drones Based on FPGA Dynamic Vision

摘要: 针对植保无人机在精准施药中“看得见但打不准”与“算得动却飞不远”的核心技术瓶颈，本研究提出一种基于FPGA动态视觉的边缘智能施药系统。该系统以动态视觉处理架构为基础，融合边缘计算能力，并优化自适应阈值算法，实现了感知–决策–执行的闭环快速响应。测试结果表明，该系统在复杂农田环境下具有低延迟、高识别准确率和低误检率的优势，系统能效也显著提升。相较于传统方案，本系统在施药精度、作业效率和续航能力方面均有明显改进，为植保无人机的智能化升级提供了可行的技术路径。

Abstract: To address the core technical bottlenecks of “seeing but not spraying accurately” and “capable of computation but limited by flight endurance” in precision pesticide application for plant protection drones, this study proposes an edge-intelligent spraying system based on FPGA dynamic vision. The system, built upon a dynamic vision processing architecture, integrates edge computing capabilities and optimizes adaptive threshold algorithms, achieving a closed-loop rapid response from percep- tion to decision-making and execution. Test results demonstrate that the system exhibits advantages of low latency, high recognition accuracy, and low false detection rates in complex farmland environments, alongside significantly improved energy efficiency. Compared to traditional solutions, this system shows marked improvements in spraying accuracy, operational efficiency, and endurance, providing a viable technical pathway for the intelligent upgrade of plant protection drones.