Urban Transport

MIT's shared ebike transforms into a self-driving trike

MIT's shared ebike transforms ...
The MIT Autonomous Bicycle, pictured here without the front hub motor
The MIT Autonomous Bicycle, pictured here without the front hub motor
View 3 Images
The MIT Autonomous Bicycle's rear wheels, in bicycle and tricycle modes
1/3
The MIT Autonomous Bicycle's rear wheels, in bicycle and tricycle modes
The MIT Autonomous Bicycle, pictured here without the front hub motor
2/3
The MIT Autonomous Bicycle, pictured here without the front hub motor
The MIT Autonomous Bicycle has already been successfully tested on the MIT campus
3/3
The MIT Autonomous Bicycle has already been successfully tested on the MIT campus
View gallery - 3 images

While bicycle-sharing systems certainly are proving to be popular, they still have a few problems, such as the limited availability of bikes. Researchers at the MIT Media Lab are working on a solution, in the form of the MIT Autonomous Bicycle.

Many bike-sharing programs incorporate multiple "docks" throughout the city, where users pick up and drop off bicycles. Sometimes, however, the closest dock may still be inconveniently far from a user's present location, or their destination. Additionally, there can end up being an overabundance of bikes at the centrally located docks, but relatively few at those in outlying areas.

Some programs instead go for a dockless approach, wherein the bicycles are simply left at their last destination. Once again, though, this tends to result in most of the bikes ending up in a few central areas. Additionally, the bikes are often parked haphazardly, getting in peoples' way. Ultimately, in both docked and dockless systems, vans are required to periodically gather up and redistribute the bikes.

That's where the MIT Autonomous Bicycle Project comes in.

Led by research scientist Kent Larson and master's candidate Naroa Coretti, it's centered around a working prototype ebike that has two side-by-side rear wheels and two motors – one of those motors powers the front wheel, and the other steers the bike by turning that wheel from side to side.

The idea is that clients would initially summon the closest available ebike to their present location, via an app. Guided by GPS and obstacle-avoidance sensors, that bike would autonomously make its way to them with its rear wheels spread apart from one another, allowing it to function as a self-driving tricycle.

The MIT Autonomous Bicycle's rear wheels, in bicycle and tricycle modes
The MIT Autonomous Bicycle's rear wheels, in bicycle and tricycle modes

Upon reaching the user's destination, though, two linear actuators would automatically pull the bicycle's rear wheels back in together, essentially converting them into one wide rear wheel. The user would then ride the bike to their destination, with the steering motor inactivated. Once they were done with it, the bike would go back into trike mode, proceeding on to its next user or the closest charging station.

The prototype has already been successfully tested on the MIT campus, and can be seen in action in the video below.

Source: MIT Media Lab

MIT Autonomous Bicycle Project

View gallery - 3 images
5 comments
paul314
Is anyone else seized by the (subjunctive) impulse to jump on one of these bikes in autonomous mode to get a free lift to wherever it's going?
Adam Jinglebells
Ok why not just keep it in trike mode and have less things to break, less weight, and less hassle?
Warren W. Weiss
The SHIFT bicycle had a similar mechanism.

http://www.designersparty.com/entry/SHIFT-bicycle-Scott-Shim
Karmudjun
You are right Warren - but in the past 15 years I haven't heard or seen a single SHIFT bicycle roaming any suburban neighborhoods or in any bicycle shops. Plus this 'design' is similar while strikingly different with a completely different approach to serving the cycling community. But glad you remember the SHIFT....I remember reading of it in 2005 Popular Science or Popular Mechanics. But never saw it out there, and when my children starting riding, I couldn't find anyone who knew of it - even though Professor Shim is in Ohio today! But the design was pre-lithium-ebike technology, and really, well before autonomous cycling technology.
ljaques
Being a retired mechanic, I see the toe-in on those two rear tires (not to mention the many degrees of camber) and I'd give those two tires a total of 1-1/2 miles before they blew out, with half the sidewall and half the tread scrubbed right off them. It's not just bad engineering, it's preventable design! RIP, MIT. Sheesh. And I wouldn't ride any bicycle with a front tire less than 26". Smaller rims will throw you right over the handlebars if you hit anything larger than a filter cigarette butt unless you physically lift the small wheel over the obstacle.