Growth economies continue to be more reliant on roadways than ever before. Over 30,000 miles of road are added yearly to the already enormous road system that exists in the United States. As roads segment habitats, animals have no option but to walk across them for food, water and companionship. In the process they end up becoming roadkill. Wherever wildlife habitat and roadways overlap, roadkill seems inevitable. These animal deaths have a direct impact on the biodiversity and dynamics of an ecosystem and roadkill poses a threat to many species that are fighting local extinction. Vehicles colliding with animals results in human fatalities, life changing injuries and extensive property damage. Current methods for preventing roadkill are primarily passive and do not utilize the animal's natural instincts. An alternative approach is actively involving the animal, warning it of an oncoming vehicle, and triggering its survival instincts.
“Ora” is a vehicle-detection/animal warning system that could be used to deter wildlife from the roadside when vehicles approach. The workflow of Ora involves detecting the headlights of a vehicle from distances of >120 ft., emitting light and/or audiovisual signals to alert animals of oncoming danger. An inertial noise filtering algorithm managed to maximize detection distance limits false positive rate. This approach was used to create a self-learning/calibrating algorithm to identify the baseline ambient light level in an area, and compare it to the light level when there is a car present. This allows for scalability and eliminates calibration when installing in varied environments.
Animals react better to predatory sounds than tonal beeps. Predatory sounds are pre-loaded by the user into Ora and can be played randomly in order to avoid habituation. Animals are more receptive to certain light wavelengths. Different light warnings (varying spectral quality) are also pre-loaded/selected by the user into Ora and turned on randomly. The system was developed to easily configure these two modalities of alerting wildlife making it very versatile in dealing with varied forms of roadkill.
Ora was tested on a 4 lane highway under night time driving conditions, with the device located on the side of the highway. Ora detected the vehicles before they exited the detection zone 100% of the time, with an average trigger distance of 244 ft. On 96.5% occasions, Ora was able to detect vehicles from distances of 60 ft or further. The majority of detections were between 120 and 240 ft. Ora worked best for high beams and larger vehicles. Halogen lights, which formed 14% of the test set, led to an average trigger distance of 157 ft.
In conclusion, Ora provides a low cost, scalable solution to mitigate roadkill. It can help reduce permanent habitat fragmentation by allowing the animals to cross when there is no danger and warn them early when there is a potential threat.