SafeTrack is a battery-powered autonomous AIoT rover designed to improve industrial safety by continuously patrolling factory floors and detecting unsafe conditions in real time.

Industrial environments often depend on manual supervision and periodic inspections, which can miss critical safety violations and early warning signs. SafeTrack was created to automate this process and provide consistent, real-time safety monitoring without human intervention.

The rover navigates autonomously using a time-of-flight LiDAR sensor for obstacle detection and safe movement through dynamic factory environments. A camera module running an edge-AI computer vision model is used to monitor personal protective equipment (PPE) compliance, such as safety helmets, directly on the device. Performing AI inference locally reduces latency, improves reliability, and protects privacy by avoiding cloud dependency.

To enhance safety beyond visual monitoring, SafeTrack integrates a thermal imaging module that enables thermal-based safety monitoring. This allows the system to detect abnormal heat patterns, overheated machinery, potential fire risks, and human presence in hazardous or restricted high-temperature zones. Ambient temperature sensors provide additional confirmation and help reduce false alerts through sensor fusion.

When unsafe conditions such as PPE violations, overheating equipment, or abnormal temperature levels are detected, SafeTrack sends real-time alerts to supervisors through wireless IoT connectivity. The alert system can escalate warnings based on severity, enabling faster response and preventive action before accidents occur.

SafeTrack is designed with a modular and scalable architecture, allowing it to be adapted for different industrial environments such as factories, warehouses, and processing plants. By combining autonomous navigation, edge AI, thermal imaging, and IoT connectivity into a single platform, SafeTrack reduces human workload, improves safety compliance, and helps prevent industrial accidents proactively.