One of the most important aspects of independent living is mobility: the freedom to go where you want, when you want. Paraplegics and others with physical disabilities may be confined to wheelchairs, but not everyone has the strength and dexterity require to power a wheelchair, or even control a powered chair. How can we provide these people with mobility solutions?

Researchers at MIT‘s Computer Science and Artificial Intelligence Laboratory (CSAIL), the National University of Singapore, and the Singapore-MIT Alliance for Research and Technology (SMART) may have an answer to this problem. Their solution is a classic example of “Give a kid a hammer, and the whole world looks like a nail.” The group had created sensors and software to drive autonomous cars, and then golf carts. So it made sense that they would try to apply the same tools to the problem of individual mobility.

It took the team less than two months to create a working prototype of an autonomous scooter. It debuted at MIT’s Open House in the spring of 2016, and more than 100 visitors were able to take rides on the device. The fact that the scooter uses the same basic software as the golf cart and car systems means that improvements and information on one platform can be readily transformed to the others. The scientists are working to add machine learning to the system, so that the computerized “drivers” will get better with experience, and then can share that experience with all the other devices using the system.

In the not-so-distant future, we may truly be able to get door-to-door service by autonomous transportation, from a scooter that can take us from our homes to an autonomous car, and eventually to another autonomous scooter at the other end of our trips. Such a system could restore freedom of movement to a large number of people who are limited by their lack of mobility.