A tomato picking robot has been designed,which can automatically plan the path to enter the greenhouse,identify and pick ripe tomatoes.The key components of intelligent collection robots include controllers,logic controllers,etc.In the process of planning the optimal harvesting path,in order to avoid potential obstacles of the harvesting robot arm,strictly preprocessed the col-lected image data,and then used a high-precision binocular vision system to accurately obtain the three-dimensional spatial position information of fruits and vegetables in the orchard through image matching and 3D reconstruction technology.On this basis,the Hough transform algorithm and deep learning algorithm are combined to automatically plan the optimal harvesting path.The preci-sion testing results and robot picking of motion trajectory indicate that the robot has a high success rate in harvesting fruits and vege-tables,which can meet the requirements of robots in fruit picking countries.
维普
万方数据