When the project was first announced in 2015, Yamaha partnered with SRI International to develop the Motobot. The goal was to create a machine that would be capable of “visualizing data about human motorcycle operation, further quantifying the relationship between rider input and machine behavior, and then using the resulting know-how to build even better vehicles.”
Yamaha had two design goals for 2017: the first was for the Motobot to reach speeds over 200 kilometers on a track, and second to defeat seven-time MotoGP champion Valentino Rossi in a timed lap around a track. The company achieved the first goal, but failed to meet the second — Rossi beat the Motobot’s time by about 30 seconds. However, the machine is improving and that’s one of Yamaha’s biggest accomplishments.
The company is hoping that advanced machine learning, of the kind which allowed a Google computer to defeat a Go world champion, will also allow the Motobot to learn the best way to navigate tracks as it makes it way across the course. The company’s ultimate goal is to design a machine which can defeat the best human opponents. The Motobot hasn’t gotten there yet, but AI and machine learning are advancing at a rapid pace, so it will likely only be a matter of time.
Beyond the design goals of improving vehicle safety and advancing the fields of robotics, Yamaha is hoping that the lessons learned during the Motobot project can eventually be applied to other vehicles such as snowmobiles and boats.
Whether or not this means we’ll eventually see the self-driving motorcycles remains to be seen. Cars make a lot of sense as self-driving machines, since there are plenty of people who find driving a necessary evil, but the people who buy motorcycles are generally more about the ride itself, so there might not be much of a market for a self-driving motorcycle.
