Researchers at MIT are taking a ‘map-less’ approach to find a way for driverless cars to operate on country roads.
Big tech companies are conducting all their testing in city environments, where they have collected detailed 3D overviews of the urban layout. In the country, however, gathering data like this isn’t possible, with endless miles of ‘roads that are unpaved, unlit or unreliably marked’. To overcome this, a team from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) has been developing a system that allows cars to get around using GPS data and sensors on the car. Using this MapLite method on remote roads in Massachusetts, the researchers found cars could ‘detect the road more than 100 feet in advance’.
“The reason this kind of ‘map-less’ approach hasn’t really been done before is because it is generally much harder to reach the same accuracy and reliability as with detailed maps,” explained CSAIL graduate student Teddy Ort. “A system like this that can navigate just with on-board sensors shows the potential of self-driving cars being able to actually handle roads beyond the small number that tech companies have mapped.”
The sensors are the key tools in this navigation model – with GPS only used to get a ‘rough idea’ of the car’s location.
“I imagine that the self-driving cars of the future will always make some use of 3D maps in urban areas,” added Ort. “But when called upon to take a trip off the beaten path, these vehicles will need to be as good as humans at driving on unfamiliar roads they have never seen before. We hope our work is a step in that direction.”
The tests were carried out using a Toyota Prius thanks to CSAIL’s partnership with the Toyota Research Institute.