0
Article ? AI-assigned paper type based on the abstract. Classification may not be perfect — flag errors using the feedback button. Tier 2 ? Original research — experimental, observational, or case-control study. Direct primary evidence. Environmental Sources Food & Water Policy & Risk Sign in to save

IoT Established Ecoplastirid Rover Integrated with Machine Learning

2025
A. Sangeerani Devi, Lesesne William G., S J Asmitha, D Bala Dharshini

Summary

This engineering project described the design of a 4WD autonomous rover ('Eco-Plastirid') with IoT sensors and deep learning capabilities for detecting and collecting microplastic debris in outdoor environments. The system integrates computer vision for plastic identification with robotic collection, targeting small-scale remediation in Indian outdoor settings.

Plastic pollution has been one of the most serious issues that irritate eco-cycles and penetrates into food chains. According to the Plastic Waste Makers Index 2019, India also stands on the top list of global largest producers, producing more than 8 million metric tons of plastic per year. In this region, the Indian government, through various activities, agreed to take on some of the following measures, such as restrictions on light-weight and one-time plastics. However, when the plastic waste disperses itself in the environment, even at the micro level, it becomes very difficult to collect it. We have conceptualized a self-driving 4WD robotic car called the Eco-Plastirid Rover, which had special properties and objectives concerned with plastic debris detection and collection. Deep learning attached to computer vision concerning vacuum-based mechanisms that would work on varying terrains would make it possible to suck both microplastic and macroplastic debris. Fine-tuning YOLOv11 as the state-of-the-art object detection model in the full view of the onboard high-resolution camera will be adopted within the main detection framework. This will ensure that the ultrasonic sensors placed on the rover will do an accurate estimation of distances that would help in dynamic path planning and the avoidance of obstacles. An efficient high vacuum suction system sucks the selected plastic waste that is then conveyed to a segregated storage compartment under realtime fill level monitoring. The system also packs a microcontroller that caters for unattended and constant operation with limited human intervention. The plastics are collected in a container, and we can monitor the level of the container. This result in effective collection of plastic which helps in the reduction of plastic pollution and reduction of labors in cleaning work. The rover gives us the perfect solution and contributes to the sustainable environment. In short, the above solution increases the rate of recovery for plastic waste while degrading the amount of man-to-man interaction within any manner of waste handling. Something about the above technology is that it can scale entirely and apply to what probably reduces impacts caused by plastic pollution. Generally, the output makes it feasible to further ecological protection and allows many waste management activities throughout the globe.

Share this paper