MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) has developed MapLite, a framework designed to allow self-driving cars to drive on roads they have never been on and without 3D maps. The system allows navigation with only GPS and sensors.
MapLite combines simple GPS data found on Google Maps with a series of sensors that observe the road conditions. A system like this that can navigate just with onboard sensors shows the potential of self-driving cars being able to actually handle roads beyond the small number that tech companies have mapped. Together, these two elements allowed the team to autonomously drive on unpaved country roads in Devens, MA, and reliably detect the road more than 100 ft (30 m) in advance.
For the testing, as part of a collaboration with the Toyota Research Institute, researchers used a Toyota Prius outfitted with a range of LiDAR and IMU sensors. MapLite uses sensors for all aspects of navigation, relying on GPS data only to obtain a rough estimate of the car’s location. Its perception sensors then generate a path to reach that point, and LiDAR is used to estimate the location of the road’s edges. The solution can accomplish this without physical road markings by making basic assumptions about the road being relatively flatter than surrounding areas.
It still has some limitations; it isn’t yet reliable enough for mountain roads, for instance, since it doesn’t account for large changes in elevation. Next, the team hopes to expand the variety of roads that the vehicle can handle. The goal is to have the system reach comparable levels of performance and reliability as mapped systems but with a much wider range.
Click to view AVT’s 2019 ACES Award Winners best in class of