An Autonomous Trash Cleaning Robot

Abstract

Object detection is one of the fundamental tasks in computer vision. A common paradigm to address this problem is to train object detectors that operate on a subimage and apply these detectors in an exhaustive manner across all locations and scales. The exhaustive search through all possible locations and scales poses a computational challenge. The main objective of this work is to design an autonomous trash cleaning robot that could be operated in a remote place. The primary action in many repetitive tasks is picking up objects and moving them to other locations. This robot was designed as a fetching robot. It was to come with a set of objects that it was designed to detect and segregate into biodegradable and nonbiodegradable, and it is able to collect these items from the environment when there are placed at random. Experimental results show that the proposed work can provide improved results than the existing machine learning algorithms.