The Design of Movable Garbage Sorting and Recycling Device

Abstract

With the progress of social productivity and the gradual improvement of people's living standards, more and more domestic waste is produced. Aiming at the problem of household domestic waste classification, an autonomous mobile waste classification collector is designed, which combines deep learning image classification with embedded technology, integrating sound source positioning, path planning and waste classification. Taking Jetson nano as the data processing center, the time delay estimation algorithm is used to obtain the location of the sound source, the lidar is used to scan the information of the surrounding environment, the SLAM algorithm is used to model the environment in two dimensions, and the path is planned for the work of the garbage collector. Taking STM32 single chip microcomputer as the control core of the motor control system, the chassis DC motor control adopts PID algorithm for closed-loop control to achieve the function of accurate positioning. After the camera collects the image information of the garbage, it uses the Yolo algorithm to detect the target and automatically classify the garbage. This design increases the convenience of household waste classification, so that each user can classify waste more quickly and easily.