Real-time pothole detection and automated refilling with auto-swift using machine learning and computer vision
| dc.contributor.author | Gowthami A -1NH21EE033 Raksha Shankar Kajagar- 1NH21EE092 Madhura N B -1NH21IS086 Mahima Ramdas Bhat- 1NH21IS088 | |
| dc.date.accessioned | 2025-01-20T04:35:04Z | |
| dc.date.available | 2025-01-20T04:35:04Z | |
| dc.date.issued | 2024 | |
| dc.description.abstract | The increasing demand for efficient road maintenance has highlighted the issue of potholes, which affect safety, vehicle performance, and transportation efficiency. Traditional pothole detection and repair methods are slow, labor-intensive, and costly. To address these challenges, this paper presents an Autonomous Pothole Detection and Repair Bot, an automated system that integrates robotics, Al, computer vision, and automated repair technology to provide real-time, cost-effective road maintenance with minimal human intervention. The system consists of three core subsystems: pothole detection, autonomous navigation, and automated repair. The detection subsystem uses advanced computer vision and deep learning models, particularly Convolutional Neural Networks (CNNs), to identify and analyze potholes in real-time. High-resolution cameras capture road surface data, which is processed by machine learning algorithms to detect and measure defects. The navigation subsystem employs SLAM algorithms and GPS for real-time mapping and path optimization. The bot's mobility platform ensures stability on uneven surfaces, while energy-efficient batteries power its operations. Once a pothole is detected, the repair subsystem activates, deploying a material dispensing and leveling mechanism. Using materials like cold mix asphalt or polymer compounds, the bot fills the pothole with precision and compacts the material to ensure durability. Surface profiling sensors validate the repair quality. An IoT architecture enables real-time monitoring and data sharing, allowing for predictive maintenance and efficient resource allocation. The system offers several advantages over traditional methods, including eliminating human labor in hazardous environments, faster repair times, and minimized waste. The modular design makes it adaptable to various road conditions, and its use of eco-friendly materials supports sustainability goals. Field trials demonstrated high detection accuracy (over 95%) and improved speed and cost-effectiveness, making the bot a promising solution for smart cities and global road infrastructure management. | |
| dc.identifier.uri | http://192.168.75.5:4000/handle/123456789/17320 | |
| dc.language.iso | en | |
| dc.publisher | NHCE | |
| dc.title | Real-time pothole detection and automated refilling with auto-swift using machine learning and computer vision | |
| dc.type | Learning Object |