RailKavach is an AI-driven railway safety system designed to prevent train collisions with obstacles, stray animals, and humans. It uses real-time object detection, multi-modal sensor fusion (CCTV, LiDAR, Infrared), and automated braking to reduce railway accidents. How It Works: AI-Powered Detection: YOLOv5 + VGG16 models detect obstacles in real-time. Multi-Sensor Fusion: LiDAR, CCTV, and infrared cameras ensure high-accuracy detection, even in fog, rain, or low-light conditions. Automated Braking: If the system detects an obstruction and no manual response occurs, RailKavach automatically stops the train. Key Features: Scalable – Works with existing railway infrastructure. Edge Computing – Low-latency AI processing. Cost-Effective – Uses affordable sensor tech. Challenges & Future Plans: Integration: While the frontend and AI model were built, seamless real-time integration remains a priority. Data Limitations: We relied on an image dataset due to the lack of Indian Railways live footage. Access to real-time data would significantly enhance detection accuracy. 🏆 Award: 1st Place – Open Innovation Track, Hackurate, Zenevia 2024. 👨💻 Team Hawks: Anusha Pundir, Sumukh Chhabra, Yash Padam, Yash Nautiyal, Paarangat Rai Sharma.