农业机器人全方位目标识别与跟踪研究——基于MIMO-OFDM信道编码

Research on Omnidirectional Target Recognition and Tracking of Agricultural Robot Based on MIMO-OFDM Channel Coding

首先介绍了MIMO-OFDM信道编码理论和在该技术上搭建的农业机器人架构,然后利用鱼眼镜头成像、标定与校正方法建立了全方位目标识别系统,并采用图像处理技术和粒子滤波法实现了全方位目标识别与跟踪系统。实验结果表明:农业机器人自主定位精度较高,目标识别与跟踪的准确率也非常好,证明了系统的有效性和可行性。

Firstly,the theory of MIMO-OFDM channel coding and the architecture of agricultural robot based on this technology are introduced.Then,an omni-directional target recognition system is established by using fisheye lens imaging,calibration and correction methods,and an omni-directional target recognition and tracking system is realized by image processing technology and particle filter method.The experimental results show that the autonomous positioning accuracy of agricultural robots is high,and the accuracy of target recognition and tracking is also very good,which proves the effectiveness and feasibility of the system.