Development of Robo Using Deep Learning for Smart City Application

This paper presents a garbage picking robot which can operate on the ground. Waste management is being given atmost importance these days. We know how tiresome repetitive cleaning is. It takes up a great deal of resources and labour. Efforts such as recycle bins and 3R's campaign aren't completely effective. This project helps in replacing the existing infra and to develop smart waste management system by using an automatic robot that can detect the garbage accurately. With the garbage recognition and automatic navigation functions, the robot can pick the garbage on the ground in places like parks or schools efficiently. Experimental results show that the garbage recognition accuracy can reach as high as 90%, and even without path planning, the navigation strategy can reach almost the same cleaning efficiency with traditional methods. Thus, the proposed robot can serve as a good assistance to relieve dustman's physical labor on garbage cleaning tasks.