Autonomous Railroad Intrusion Detection System with UAV imagery

Reference #: 1760

The University of South Carolina is offering licensing opportunities for Autonomous Railroad Intrusion Detection System with UAV imagery.

Background:

Accidental intrusions at railway crossings pose a significant threat to railroad safety. According to the Federal Railroad Administration (FRA), 2,147 railroad crossing accidents were reported in 2021, leading to 236 deaths and 666 injuries. Over the years, substantial efforts have been made to improve rail crossing safety through measures such as gated barriers, enhanced road signs, traffic warning lights, and surveillance cameras. However, many accidents still occur due to unpredictable factors such as unauthorized trespassing, track obstructions, and vehicles fouling the tracks. Traditional surveillance systems often fail to detect obstacles in real time and provide timely alerts to both train operators and road users. To address this issue, there is an urgent need for practical solutions that can detect track intrusions and mitigate the risks of potential accidents. Advances in unmanned aerial vehicles (UAVs), including autonomous drones, make it possible to develop an intelligent track intrusion detection and track integrity monitoring system.

Invention Description:

This invention is an autonomous railroad track intrusion detection system that integrates an unmanned aerial vehicle (UAV) as a mobile platform with edge-computing devices for real-time object detection, tracking, and alerting. The system leverages artificial intelligence (AI) and computer vision technologies to enhance operational efficiency and safety. The UAV is equipped with an AI-assisted flight control algorithm that enables autonomous navigation along the railroad track. By detecting and following track features, the UAV maintains a precise flight path for continuous monitoring. An AI model deployed on the edge-computing device processes real-time visual data to detect, classify, and track objects along the railroad track, identifying anomalies and potential intrusions using advanced computer vision techniques.

The complete system is designed to:

1) Autonomously navigate and patrol along the railroad track;

2) Detect and track obstacles or unauthorized intrusions in real-time; and

3) Automatically return to the base station for battery recharging.

Potential Applications:

Railroad industry

Advantages and Benefits:

Beneficial characteristics of the invention include:

1. Fully Autonomous Navigation and Patrol – The system can independently fly along railroad tracks, eliminating the need for a human pilot or track inspector. This reduces labor costs, enhances efficiency, and enables continuous monitoring without human intervention.
2. Access to Remote and Hard-to-Reach Areas – The UAV can reach locations that are difficult, dangerous, or impractical for human inspectors, such as rugged terrain, remote areas, and elevated tracks. This significantly reduces risks and injuries associated with manual track inspections in hazardous environments.
3. Real-Time Obstacle and Intrusion Detection – Using advanced artificial intelligence (AI) and computer vision, the system can detect track obstructions, unauthorized intrusions, or potential hazards instantly. As a fully automated system, it eliminates human errors, fatigue, and delays in response time, improving railway safety.
4. Automated Data Backup, Record Keeping, and Information Sharing – The system automatically stores and organizes surveillance data, ensuring a continuous record of track conditions. It can also share real-time alerts and reports with relevant stakeholders, including railroad operators, maintenance teams, and emergency responders, facilitating prompt decision-making and preventive actions.
5. Automated Return and Battery Recharging – Once the mission is complete or the battery reaches a critical level, the UAV autonomously returns to its base station for recharging. This ensures uninterrupted operation with minimal human intervention, enabling continuous, long-term monitoring of railroad tracks.