CornerCameras Technology Helps Cars to See Around Corners

One problem with self-driving cars is that they use sensors to detect their surroundings, but these sensors can only tell them about what is in their line of sight. The technology is not very good at detecting objects that are around the corner. A human driver is good at this and depends a lot on his peripheral vision. The team at the Computer Science and Artificial Intelligence Laboratory (also known as CSAIL) research laboratory at MIT has developed an innovative camera technology called CornerCameras technology. CSAIL is the largest research laboratory at MIT and one of the key centers in the globe for research in this area.

The technology they developed uses small variations in lighting to understand what might be around the corner and is based on the concept that objects reflect a small amount of light around walls, thus creating a “shadow”. Such “shadows” can be used to recreate images of objects around the corner by capturing video of the floor and tracking the shifting colors.

The CornerCameras system is surprisingly dependable and has the capability to work in a wide range of lighting conditions, both indoors and outdoors and even in inclement weather like rain. Another benefit of this technology is that it does not call for the use of expensive components like lasers. This can be easily mounted on an automated vehicle and can give good results.

The team at MIT is still ironing out the kinks. One problem is that the camera cannot work when there is no light in either the visible or hidden scenes. The system is also easily confused when the lighting changes constantly. While it might take a while before the technology has been perfected and can be safely used on automated vehicles, it has already been tested on a wheelchair and could be used for making piloted vehicles safer as it could warn you when someone runs out into the street.