With the rapid development of modern livestock and poultry industry,the demand for mechanization and automation of breeding technology in livestock and poultry farms is constantly increasing.The precise feeding technology of pigeons can not only effectively improve feed utilization and reduce labor costs,but also improve production efficiency and economic benefits.This article aims to design an intelligent feeding cart for pigeon breeding with the goal of intelligent and precise feeding.Firstly,introduce the hardware of the intelligent feeding cart for pigeon breeding;Then,use the YOLOv7 deep learning model to identify the number of pigeons in three different height pigeon cages;Next,develop ultrasonic detection function for remaining feed in the silo,infrared tracking function for path tracking and navigation;Finally,experimental analysis was conducted on the feeding efficiency,operational accuracy,and optimal feeding distance of the intelligent feeding cart for pigeon breeding.The results showed that under normal indoor lighting,the accuracy of identifying pigeons and feeding troughs was 93.36%,the average time for detection and feeding was 2 seconds,and the feed rate was below 1%,effectively improving the operational efficiency of the pigeon breeding farm.