Low-Resolution Maps for Self-Driving Cars
Conventional wisdom is that safe and effective autonomous vehicle navigation requires extremely detailed 3D maps in order to guide cars efficiently and safely through all of the complexities and variables of modern roadways. This places a significant resource burden on the companies developing self-driving cars, as they must carefully map any locations in which their cars will operate.
This might not be the case for much longer, according to new research out of MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL). Led by Daniel Rus, the lab has released MapLite, a framework that allows self-driving cars to drive on new roads using GPS and LiDAR; and more importantly without existing detailed 3-D maps.
“Our minimalist approach to mapping enables autonomous driving on country roads using local appearance and semantic features such as the presence of a parking spot or a side road. 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,” said graduate student Teddy Ort.
Read more from MIT News - Here.