Impro­ving Rail Safe­ty with Tra­in Moni­toring Tech­no­logy and Infrast­ruc­tu­re Invest­ments

Rail safety has become an urgent priority in the United States, underscored by the current Administration's recent allocation of over $1.1 billion through the Railroad Crossing Elimination (RCE) Program. This initiative, part of the Infrastructure Investment and Jobs Act, aims to address the dangers associated with at-grade railroad crossings by funding crucial infrastructure projects, including track relocations, grade separations and the installation of advanced safety systems. These efforts are designed to significantly reduce the risks of accidents and fatalities at these crossings.

While these infrastructure improvements are essential, they must be complemented by modern technological solutions, such as comprehensive train monitoring applications. One such application, developed by Noema, plays a critical role in enhancing the safety and efficiency of rail networks by leveraging real-time data and predictive analytics.

The Need for Train Monitoring Applications

Rail crossings across the U.S. vary greatly in terms of traffic, location and the condition of surrounding infrastructure. Many of these crossings are in high-traffic areas or regions with aging infrastructure, increasing the risk of accidents. The RCE Program aims to mitigate these risks through physical improvements, but technology like Noema's Train Monitoring Application can provide an additional layer of safety and operational efficiency.

Below are key features and impacts to be considered for train monitoring applications to address the complexities and challenges of modern rail networks:

• Real-Time Monitoring: The application tracks train movements using camera sensor data which, when combined with GPS location, provide localized real-time updates that allow for the immediate detection of anomalies. This capability is particularly crucial at high-risk crossings, where quick responses can prevent accidents.

• Predictive Analytics: By analyzing historical and current data, an application can address potential hazards, such as unexpected blockages at crossings. This predictive power enables operators to take preemptive actions, such as adjusting train speeds or rerouting trains to avoid accidents. 

• Automated Alerts: The system issues automated alerts to operators, drivers, and local authorities when potential risks are detected. These alerts facilitate timely interventions, such as lowering crossing gates earlier or dispatching emergency services, thereby reducing the likelihood of collisions.

• Integration with Infrastructure: Application must seamlessly integrate with existing rail infrastructure, including signals, crossing gates, and communication systems. This ensures that all elements of the rail network work together harmoniously, enhancing the overall safety and efficiency of the system. 

• Data Analytics and Reporting: The application should collect large volumes of data and the ability to be customized to generate reports that help operators understand safety trends, identify areas for improvement, and comply with regulatory requirements. This data-driven approach leads to better decision-making and resource allocation.

• Scalability and Customization: Designed to be scalable, the train monitoring application must be suitable for both small regional railways and large national networks. It is also customizable, allowing operators to tailor the application to their specific needs, whether for basic monitoring or more advanced features like AI-driven predictions.

• Extension of Analytics like Hazard Detection: The application should include or be customizable to include: 

  • Crossing gate state (open/closed) and malfunction detection
  • Realtime Fire and Smoke hazard detection, both on the railway, trains, passing cars and surroundings
  • Railway Flooding from rain, nearby rivers or sea
  • Snow storms blocking the passage
  • Human or Car railway unauthorized crossing

The combination of government-funded infrastructure improvements and cutting-edge technology solutions, such as Noema's Train Monitoring Application, represents a comprehensive approach to rail safety. The real-time data, predictive analytics, and automated alerts provided by Noema’s system enhance the effectiveness of physical safety measures, ensuring that the significant investments made today will lead to long-term safety and operational benefits.

As rail networks continue to evolve, integrating advanced monitoring systems will be essential to maintaining and improving safety standards, ensuring that both public and private investments in rail infrastructure protect lives and enhance the efficiency of rail transport across the United States.

What Differentiates Noema’s Train Monitoring Applications? 

Noema’s application utilizes state-of-the-art edge AI-based Computer Vision to accurately detect and monitor trains in the field of view of a simple IP camera or a set of cameras. 

It collects pixel-wise information, meaning we know how many trains are in the field of view, their relative size, their speed, their path, and so on. And then, from the speed information, we can derive other data points like:

• “is the train slowing down?”
• "is the train stopped?”
• "is the train accelerating?”

During the daytime, it works with visual streams (RGB). At night, it can also work with visual streams, provided there is enough lighting (such as street lights), or it can work with infrared streams, in case external lighting is not available in remote locations. 

Noema’s Training Monitoring application runs on the edge, meaning:

• it packs lightweight AI algorithms, which allows us to parallelize more AI tasks into the same hardware. We can do things like hazard detection (fire, smoke, flooding, snow), unauthorized human or vehicle crossing, or any custom analytics requested by the customer
• it requires low power and low bandwidth, and can therefore be installed in the most remote locations, guaranteeing more data points and allowing for better predictions
• it is very much hardware agnostic and can be installed on existing camera systems
• local network or internet are both supported

Although it does not require any other sensors apart from a camera, it supports integration with external sensors via GPIO, REST, MQTT and others. The underlying algorithm logic can be extended with data from existing microphones, crossing gate status sensors, rain detectors and others, to produce more accurate predictions and richer reports. Through the same protocols, it can be integrated with control loops such as the crossing gate controlling mechanisms, sound alarms, or even send alarms directly via email, SMS, or REST calls over the internet.

The integration of advanced monitoring systems like Noema's Train Monitoring Application are essential for modern rail safety. By leveraging real-time data, predictive analytics, and automated alerts, Noema's technology provides a crucial layer of safety that complements physical enhancements. This comprehensive approach ensures that both public and private investments in rail infrastructure translate into long-term safety, operational efficiency, and protection of lives across the U.S. rail networks. As rail systems continue to evolve, Noema's cutting-edge solutions will play a pivotal role in maintaining and advancing safety standards.