SegNet Allows Self Driving Cars Learn to Self while driving on the road

12.06
SegNet Allows Self Driving Cars Learn to Self while driving on the road -

The major car manufacturers are competing to come up with a technology that allows vehicles to drive. Some manufacturers use sensors and some use LiDAR technology for self driving vehicles to detect the environment around them and make a decision on that basis. An accurate virtual card also takes an important role to allow the autonomous system knows what path he must take to bring its passengers to their destination.

A team from the University of Cambridge invented a new system called SegNet which is designed to analyze the road and its various features, such as street signs, road markers, people and even the sky. Based on the official version, the system takes a picture of the street in RGB mode and accordingly in real time class. There are a total of 12 different categories, such as roads, traffic signs, pedestrians, buildings and cyclists. In addition to this, it can also differentiate between the shadow and the night lighting environment from time.

SegNet is also able to allow the vehicle to detect an object and assess its position and orientation of a few meters and a few degrees. On top of that, it does not require a wireless connection to analyze and report a position.

Professor Roberto Cipolla, the leader of the research said that in the short term, they were more likely to see this kind of system on a home robot such as a robotic vacuum cleaner, for example. He also added that it would take time before the drivers could fully trust an autonomous car, but more efficiently and accurately they could do these technologies, the more they were to widespread adoption of driverless cars and other types of autonomous robotic.

Previous
Next Post »
0 Komentar