Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{195884,
        author = {Mr. Mari Rajan S and Priyangaa Thirumalai and Ramesh S and Lokesh P and Pradeep K},
        title = {AI-Based Automatic Railway Crossing Gate Control System},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {1695-1701},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195884},
        abstract = {Inadequate gate operation at railroad level crossings can result in severe collisions, making them a high-risk location for accidents in any transportation network. This paper describes a system that uses artificial intelligence to control railway crossings using AI, eliminating manual intervention through intelligent real-time automation. An ESP32-CAM module employs Edge Impulse to train a FOMO object detection model and identify approaching trains on the device. Upon detection with confidence 0.50, the module publishes "ON" to an Adafruit IO MQTT feed over Wi-Fi, while otherwise it publishes "OFF." A Raspberry Pi Pico W subscribes to this feed and responds by rotating a servo motor to close the barrier, which activates essentially serves as a warning. Once the train has left, the gate automatically reopens. After a repeated run of testing based on constructing ten prototypes of the same scale, the detection confidence was 0.50–0.85 with almost zero inference latency and the end-to-end gate response under 500ms. The outcomes demonstrate that inexpensive AI hardware can deliver dependable, fixed railway crossing automation. Contains information on Edge AI, ESP32-CAM, FOMO Object Discovery, IoT, MQTT, Railway Safety and Raspberry Pi Pico. Were the index terms.},
        keywords = {},
        month = {April},
        }