Self-driving cars usually need pre-planned road networks to help them find their way and travel. They also want a detailed 3D map that would lead them to their safe destination. Several companies have been testing new methods to make navigational self-driving cars for unmapped roads. Therefore, a team of researchers at MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) developed a new system to navigate cars.
The system is called MapLite, it allows the cars to drive on the roads they had never visited. The system works without the help of 3D map because it already been infused with GPS data, the same data you find on Google maps. Plus it has the sensors that will determine the condition of the roads.
MIT describes how this system will work:
MapLite uses sensors for all aspects of navigation, relying on GPS data only to obtain a rough estimate of the car’s location. The system first sets both a final destination and what researchers call a “local navigation goal,” which has to be within view of the car. Its perception sensors then generate a path to get to that point, using LIDAR to estimate the location of the road’s edges. MapLite can do this without physical road markings by making basic assumptions about how the road will be relatively more flat than the surrounding areas.
The researchers use Toyota Prius car in collaboration with the company, the cars are integrated with LIDAR and IMU sensors. The researchers then test the self-driving cars at unmapped roads of Devens and Massachusetts, interestingly the cars not only planned the route but also anticipated 100 yards of the route ahead.
According to Daniela Rus the director of MIT’s CSAIL,“The cars use these maps to know where they are and what to do in the presence of new obstacles like pedestrians and other cars. The need for dense 3-D maps limits the places where self-driving cars can operate.”
The team will also develop a system that is “parameterized,” and is able to tell you about the type of roads, the turns, and intersections. MapLite is a different technology you don’t need any maps for it everything will function under registered data, ultimately you be the one who’s asking questions about paths, directions, and intersections from the computer.