
3 items • 3 images
Aerium Train Tracks Drone Monitoring combines aerial AI vision, real-time train approach detection, track worker safety alerts, and centralized drone fleet management in a unified interface. It delivers live aerial camera streaming, automatically identifies approaching trains along active rail segments, calculates proximity to designated work zones, and broadcasts instant alerts to field personnel and central operations. The solution ensures proactive hazard prevention and continuous track monitoring.
Software and System Development
IoT Middleware Development
3 total items
August 26, 2026
Ensuring the safety of railway maintenance personnel is one of the most critical challenges in rail infrastructure management. Aerium Train Tracks Drone Monitoring is built to address this challenge by leveraging autonomous drone aerial surveillance and edge AI detection to safeguard workers on active tracks.
By connecting directly to drone camera payloads and aerial video streams, the system monitors rail corridors in real time. Instead of relying solely on manual lookouts or timetable schedules, AI processing actively scans the tracks ahead to identify approaching trains.
The platform delivers low-latency live video streaming from airborne drones while continuously analyzing track sections for incoming rail traffic.
It detects crucial operational information, such as:
This continuous visual coverage allows safety supervisors to detect approaching trains well in advance and initiate timely warning protocols.
Aerium integrates with field alert systems and central operations via secure communication protocols such as MQTT and WebSockets. This ensures critical warnings reach personnel without delay.
The system automatically broadcasts:
This real-time connectivity helps bridge the gap between aerial surveillance and immediate ground-level action.
In addition to railway and worker monitoring, the application supervises the operational status of deployed drones and base stations.
It provides key diagnostic information, including:
Alerts are triggered immediately if a drone encounters low battery levels, signal degradation, or unfavorable flight conditions.
The software provides a unified desktop and web control panel organized into core operational modules.
Displays live aerial video feeds, train tracking overlays, worker zone locations, and instant alert notifications.
Presents real-time telemetry, battery health, connection quality, and compute performance for active drones.
Allows administrators to configure detection zones, safety alert thresholds, worker geofences, and communication channels.
The system supports remote flight command dispatch, allowing authorized operators to trigger automated patrol routes, direct drones to inspect specific track sections, or command return-to-home protocols. This reduces the complexity of drone operations and ensures consistent aerial coverage across remote or difficult-to-reach railway segments.
An AI-powered desktop surveillance application developed with Python and Qt, combining human detection, thermal detection, and automated alarm triggers to provide real-time monitoring and alerts for critical safety scenarios.

i-Cross Edge AI Control Panel is a smart pedestrian monitoring application designed to run on an NVIDIA Jetson Nano. It uses an AI-powered camera system to detect people, monitor pedestrian movement, and provide real-time information about the crossing area. The application also helps operators monitor the device's condition and receive important alerts, making it suitable for smart city, pedestrian safety, and traffic monitoring applications.

In collaboration with Betamotor Italy, we developed an innovative Android application that enhances the riding experience by enabling seamless communication with Betamotor bike ECUs via Bluetooth. This project was aimed at providing riders and mechanics with a powerful tool for real-time monitoring, health checking, and tuning of their motorcycles.