An Integrated Self-Diagnostic and Communication System for Autonomous Trash Can Robots

Abstract

This paper introduces two interrelated supporting features of an intelligent trash can robot systems: an AI (Artificial Intelligent)-driven self-diagnostic system and an IoT (Internet of Things) communication system for human-robot collaboration. The self-diagnostic system assesses the robot's operational status based on internal and external parameters. When critical conditions arise, such as a full trash can, low battery, or operational failure, the robot communicates with personnel via a Telegram bot, providing real-time status updates. Experimental results demonstrate that the Naive Bayes-based self-diagnostic system achieves an accuracy of 96.67% and an F1 score of 0.95. Furthermore, a developed communication system based on MySQL database significantly reduces data transmission time from 5 seconds using Wemos D1 mini to under 2 seconds.