Taipei:The Taiwan Roadkill Observation Network (TaiRON) has proposed an innovative use of artificial intelligence to simulate and predict road sections at high risk for wildlife-vehicle collisions. This initiative also includes the establishment of an early warning network aimed at preventing roadkill and related traffic accidents.
According to Focus Taiwan, over the past eight years, an average of 3,000 traffic accidents per year have involved animals suddenly running onto roads. Motorcycles were involved in 95 percent of these cases, resulting in a total of 25,345 injuries. Authorities estimate the direct economic losses from such accidents range from NT$800 million (US$24.55 million) to NT$6.4 billion annually. TaiRON noted that the actual number of accidents is likely higher, as available statistics cover only major cases, with estimates of collisions involving medium- to large-sized mammals alone numbering between 500,000 and 1 million annually.
The proposal aims to proactively identify high-risk road sections, including those not previously recognized, by integrating existing datasets. These datasets include TaiRON's records of animal deaths, wildlife activity data from the Taiwan Biodiversity Network, public reports of injured or dead wild animals, and Ministry of Transportation and Communications traffic accident statistics. The data would be cross-referenced with road network maps, speed limits, traffic volumes, and other relevant information to identify conditions associated with frequent roadkill. AI would then simulate and predict high-risk areas.
Additionally, TaiRON proposed compiling a nationwide inventory of changeable message signs linked to a real-time warning system with heat and light sensors to alert road users of animals crossing or approaching roads. Initially, the prediction system would focus on the pangolin, with plans to expand to other wildlife such as leopard cats, endangered land crabs, and frogs.
A low-cost, solar-powered warning device developed by TaiRON, which works in low light and uses infrared and light sensors, will be tested on roads managed by the New Taipei City government. This proposal has been submitted to the 2026 Presidential Hackathon and has advanced to the final round in November. Winners will receive government support to implement their initiatives.
TaiRON, a citizen science initiative launched in 2011 by the Ministry of Agriculture's Taiwan Biodiversity Research Institute, has documented wildlife deaths associated with road infrastructure. The network aims to reduce threats to wildlife, improve road safety, and prevent collisions, with over 36,000 members contributing more than 330,000 records of animal deaths involving over 600 terrestrial vertebrate species.