Autonomous Cars are Having Trouble With Animals

author: Vineeth Joel Patel   

Self-driving cars will definitely help drivers in cities. In fact, that's why the majority of companies and automakers are testing their autonomous vehicles in densely populated areas, like Pittsburgh, Detroit, and Silicon Valley. In cities, driverless cars will be required to detect other vehicles, pedestrians, and ever-changing road patterns. 

But moving away from cities, things, presumably, get a lot easier for self-driving vehicles. Road patterns in suburban areas don't change as much as they do in cities, pedestrians, aren't as common, and the overall traffic pattern tends to be much more subdued. But autonomous vehicles will also have to face one more item in suburban areas – animals. 

Kangaroos Prove To Be Tricky

According to a report by Marketplace, animals, especially kangaroos, are proving to be extremely difficult for driverless vehicles to deal with. While automakers and tech companies have figured out a way for driverless cars to avoid getting into an accident with moose, elk, and deer, kangaroos still pose a major issue for autonomous vehicles. 

As the outlet reports, kangaroos account for 80 percent of animal-related accidents in Australia. Unlike other animals, kangaroos are extremely quick and agile, making it hard for self-driving vehicles to detect and react to the creatures. Marketplace spoke with Pedro Domingos, a computer scientist at the University of Washington, in regard to why autonomous vehicles are having a hard time detecting kangaroos. 

When questioned on why driverless systems can detect elk, moose, and deer, but not kangaroos, Domingos believes that the problem lies with awareness. "They're not terrible partly because people saw this was a problem and they've worked hard on it to detect these animals," said Domingos. "Deer and moose, they're bigger and they move slower." 

Earlier this year in February, Volvo rolled out its new Pilot Assist II system that is packed with a lot of new features aimed at reducing the amount of animal-involved accidents. Other automakers, like Mercedes-Benz and Cadillac have similar systems in place, as well. 

Domingos then explained why driverless cars have so much trouble with kangaroos. "They can jump quickly across the road and in front of the car," he said. "They're low. And they also change shape as they hop more than a deer does, which makes it harder for even a state of the art computer vision system to understand that yes, this is a kangaroos and not a wind-blown plastic bag." 

Bespoke Technology For Specific Locations

While drivers in North American won't have to worry about kangaroos crossing the road, the situation would change how automakers and technology companies would change autonomous systems for specific regions. 

"I think cars that get deployed in different countries and environments will have to be trained for those environments," said Domingos. "Having said that, the reason we have self-driving cars today is because of machine learning. It's not because somebody programmed into the car what to do. In fact, nobody actually knows how to program a car to drive. The car watches the person drive and learns to do the same." 

If self-driving cars are having a hard time dealing with kangaroos, it proves that the systems are still far away from being perfect. While autonomous vehicles can avoid accidents with other vehicles and pedestrians, animals still seem to throw driverless vehicles for a loop. Hopefully, automakers and tech companies can improve the technology behind detecting animals. 

via: Marketplace