A year after its implementation, the country’s first artificial intelligence (AI) and machine learning-enabled surveillance system, designed to prevent elephant deaths on railway tracks, has proven successful along the busy interstate route between Palakkad in Kerala and Coimbatore in Tamil Nadu.
Installed by the Coimbatore Forest Division, the facility detects elephant movement on the busy tracks that pass through the Walayar-Madukarai forests around the clock and alerts train operators instantly to avoid collisions.
Since its inauguration in February 2024, the AI surveillance system has effectively safeguarded the traditional migratory path in the Walayar-Madukkarai forest range. Notably, no wild elephants have been killed on the two railway tracks, marking a significant success for elephant safety.
The tracks connecting Kerala to Tamil Nadu and other states often turn into dangerous areas for elephants moving through the critical Western Ghats region and its interconnected forests.
The growing elephant population, disruptions in migration corridors, increased development activities along animal migratory paths, changes in land use patterns and agricultural practices, and human pressure have all contributed to a rise in human-elephant conflict in the region.
The elephants in the Coimbatore Forest Division migrate seasonally from the Nilgiris and Sathyamangalam to forest areas in Kerala.
“Specific locations that elephants prefer as retreats during the rainy season include the interstate regions such as Walayar, Boluvampatty, the Anaikatti Reserve Forest, Gopinari Reserve Forest, Hulical, the Jaccanari slopes, the Nilgiris Eastern Slope Reserve Forest, Soolakkarai, Singapathy, and the Iruttu Pallam blocks,” stated conservation activist S Guruvayurappan.
Between 2021 and 2023, elephants in the Coimbatore forest division strayed out approximately 9,000 times.
“One of the major conflict issues in the Coimbatore division involves elephants crossing the railway tracks in the Madukkarai range, which has resulted in train accidents. Two railway tracks run through the Soolakarai beat and Boluvampatti Block-I Reserved Forests within the Madukkarai Range. This forest area shares a boundary with the Kerala forests along the Walayar river and is frequently visited by elephants. Unfortunately, since 2008, 11 elephants, including young calves and juveniles, have died due to collisions with trains,” pointed out Guruvayurappan.
Despite the forest department’s best efforts, elephant incidents could not be significantly reduced. As a solution, a proposal was made to implement an AI-enabled autonomous surveillance system to track elephants and prevent accidents.
A field investigation identified a seven-kilometre section of the railway track as the most vulnerable area. In response, the Tamil Nadu government allocated Rs 7.24 crore to install an artificial intelligence system.
Supriya Sahu, the Additional Chief Secretary for Environment, Climate Change, and Forests in Tamil Nadu, informed that a dozen AI-enabled cameras were installed in the area as part of a pilot project in November 2020 by the then Forest Minister, M Mathivanan, who officially inaugurated the installation in February 2024.
“Now, a year after commissioning, the results are stunning: zero elephant accidents, 5,011 AI-generated alerts, and 2,500 safe elephant crossings,” informed Sahu.
“As part of the project, AI-powered thermal cameras have been installed on 12 towers. These cameras are monitored by a command centre staffed by local tribal youth, allowing for round-the-clock surveillance of railway tracks. They provide real-time alerts to locomotive drivers and patrol teams. Additionally, elephants extensively use two underpasses built by the railway. This initiative is now becoming a model for technology-driven wildlife conservation across the country,” she said.
Now a model to emulate, the project’s advanced control and command centre is about 1 km from the Walayar interstate border.
In 2021, the Madras High Court directed the forest department and railways to take measures to prevent elephant deaths on railway tracks. According to Sahu, developing an AI-based system posed challenges because elephants are highly intelligent and have learned to adapt to traditional control measures such as trenches and solar fencing.
Sahu noted that approximately 130 trains pass along the A and B railway lines daily, and nearly 1,000 elephant crossings are reported on these tracks annually. The forest department collaborated closely with the railways to implement this project.
Project manager Ashish Rajput stated that the AI system’s cameras, similar to those used by the Indian Army along the nation’s borders, are designed to detect humans near railway lines. This feature helps prevent human accidents and serves as an early warning system for potential human-elephant conflicts. When elephants are spotted within 100 feet (ca. 30 m) of the railway track, alerts are sent to forest and railway officials. These officials then work together to slow down trains and redirect the elephants to prevent collisions. Four personnel continuously monitor the system from a control room near the railway track.
Forest personnel previously conducted regular patrols along railway tracks to monitor elephant activities. Sahu noted that this method had limitations, making it challenging to ensure complete safety for elephants from accidents.
“Alerts are sent if any animal, not just elephants, is found near the track,” said R Manikandan, who works in the control room.
Tamil Nadu’s announcement to expand the AI surveillance system to four vulnerable areas, including Dharmapuri and Hosur, is a promising step towards enhancing wildlife safety. This expansion plan instills hope for the potential impact of the system.
“Technological interventions significantly reduce train-related accidents involving wildlife,” Sahu stated.