Wildlife road accidents threaten endangered species, but roadkill incidents also jeopardize drivers and passengers, and may oblige road operators to disburse costly compensation. Considering that domestic animals such as horses, cattle and dogs are just as likely to be victims, the scale of the problem is apparent. More than 23,000 road accidents involving users and animals were recorded in the Brazilian state of São Paulo alone between 2005 and 2013.
Measures to mitigate wildlife-vehicles collisions drive biologist Fernanda Delborgo Abra, founder of ViaFauna, a road mitigation firm. Together with two partners—Mariane Rodrigues Biz Silva e Paula Ribeiro Prist—she started the concept for Brazil's first-ever electronic roadside animal detection system. "The scientific literature shows animal detection systems can reduce the number of collisions by as much as 90 percent," says Abra.
Christened as Passa-Bicho, the prototype was developed with support from the São Paulo Research Foundation's Innovative Research in Small Business (PIPE) Program. The system comprises a pair of motion sensors (transmitter and receiver) affixed to short poles similar to those used for speed traps and installed 100 m apart. "Each pair of sensors covers a roadkill hotspot," Abra explains.
The transmitter sends the receiver a beam of infrared light that is invisible to humans and other vertebrates. When the beam is interrupted by an animal, the sensor transmits a signal to the pole, which in turn transmits the information via radio, activating an electronic message panel or, in a simpler version, a revolving beacon light on top of an animal crossing sign.
"Our detection system warns drivers hundreds of meters or even a kilometer or two ahead of an actual animal crossing, giving them time to take precautions," says Abra. "This system is far more effective than a mere sign warning that wildlife may cross the road. When drivers see static warning signs, they never know when animals will actually cross, and so they tend not to pay too much attention."
Wild and domestic animals of medium-to-large size (over 3 kg) are targeted by ViaFauna's detection system because of their detectability and impact on road safety.
During phase 1 studies of PIPE, between June 2016 and April 2017, the researchers made important modifications to a Passa-Bicho prototype. Initially, the system would emit visible light to help users see an animal as it crossed the road, but that was discarded. "Our studies led us to conclude that a floodlight would attract insects and their predators, and even scare some wildlife species away," Abra recalls.
After phase 1, the firm developed a functional prototype powered by solar panels and including data loggers. "Currently, researchers only know how many animals are killed on a road. Our system will show how many cross the road successfully, contributing to studies of animal movement dynamics," Abra says.
The biologist explains that ViaFauna opted for the prevalent system in European countries, simpler than those employed in the United States. On one hand, American models are more sophisticated for their use of thermal imaging cameras and recognition software capable of decision making, registering every animal that crosses the road. And they only warn road users when the animal is a safety hazard.
The next step will be to test the prototype at the University of São Paulo's Luiz de Queiroz College of Agriculture (ESALQ-USP) and see how it withstands weathering. "We also need to make several changes to the prototype to make its design appear more 'market-friendly,'" Abra explains.
Since phase 1, Abra has pursued the information she needs to enable dialogue with experts in the system's electronics. For phase 2, the team will be even more multidisciplinary as ViaFauna is joined by Trapa Câmera, a firm that specializes in wildlife camera traps and other environmental monitoring devices, and Hoobox Robotic, a robotics and artificial intelligence startup also supported by PIPE. Both firms will participate in tests of ViaFauna's prototype. "Today's professionals have to be multidisciplinary. You can't stay within the confines of your initial training," Abra says.
Explore further: When cars and wildlife collide: Virtual reality could prevent real-life road accidents