A classic Nintendo racing game has become the driving force behind the future of self-driving vehicles.
Videos by Wide Open Country
Super Mario Kart sold over 8 million copies following its June 1992 release. The groundbreaking installment kick-started one of the most successful video game franchises.
Now, the smash 1992 SNES racing game is being used to help refine algorithms for autonomous cars.
"It's like our Tamagochis back in the day," associate professor at the University of Maryland, Dr. Mumu Xu said. "We're trying to provide mathematical guardrails to show that AI systems can be safe."
Xu studies aerospace engineering and is conducting the study with her students.
Funded by the U.S. Naval Air Warfare Center Aircraft Division, the researchers are using artificial intelligence (AI) to train a computer to "play" Mario Kart on its own.
In this case, Mario is programmed to finish a lap as quickly as possible while still abiding by the laws of physics and, most importantly, avoiding risk.
"We give positive points for finishing the lap, negative points for going off the road, negative points for going backwards," Xu explained. "[Mario] will over time learn what gives him more points and what doesn't. And he eventually learns that if you finish a lap you get the most amount of points."
Ultimately, the research team hopes to stymie the narrative that autonomous cars are dangerous.
UMD Students Use Mario Kart to Revolutionize Autonomous Cars
Completion is merely a part of the process. The UMD researchers aim for safety above all.
"Right now, AI is such a black box that there's no formal way to show that something is safe. And so, it's a little bit scary if you're thinking about putting these cars on the roadway. I can't guarantee that it's going to perform well every single time," Xu said.
With the help of the Nintendo icon Mario, Xu and her students plan to make autonomous driving more seamless.
Like with any great invention, there were a few hiccups along the way.
"It started out where he was hopping and just wouldn't move at all. Because he didn't want to get penalized," Xu explained. "He 'reward hacked' and just basically made loops over and over again to checkpoint one and would come back to the start and come back again."
Eventually, Mario drove the full length of the track. Xu says it took six months.
Now, Xu and her research team can pull meaningful data from his laps. Soon, the findings could make roads safer for autonomous cars.
