The self-driving car field has been growing day by day; Drive.ai which is mainly focused on the field is using deep learning in a way to teach driverless vehicles to communicate with humans, an advanced form of artificial intelligence.
Drive.ai which was formed by graduated student’s under the Artificial Intelligence Lab of Stanford University, newly unveiled the company and their plans, in addition and cooperation of the former GM (General Motors) director Steve Girsky as one of the board members for the company.
The company rose funding worth $12 million from a handful of strategic investors and a venture capital company. The company was keeping quiet until April, when the California government awarded the startup with a testing license for self-driving vehicles in the state, this drove suspicion into many who were wondering what Drive.ai was working on as it was never disclosed.
After much wait it has been unveiled that Drive.ai was working on an uncontrolled driving system that utilizes deep learning in every type of different form. The self-driving cars they have made will be given a variety of different situations and objects to test how they perform, with their system and experts deducing the knowledge that it gains from these tests to decide on what to do when it encounters new situations.
The developing technology Drive.ai is working on will still require an ample amount of time of testing to take basic information, that explains the application for a license in the state of California, but Drive.ai president and co-founder Carol Reiley believes that this type of testing will help and grow in the development of the startup’s main system as it will be able to encounter more unique cases.
From Reiley’s perspective, Drive.ai is using deep learning in its self-driving technology not just mainly for functions for example detecting objects, but to make the vehicles independent on what would be the safest for the cars passengers and the cars and pedestrians that may be on the road.
“We think that deep learning is the definitely the key to driving because there are so many different edge cases,” Reiley added, saying that the startup has seen many of cases. For example a very large number of these situations would make it impossible to write a rulebook that a self-driving car could follow in an ease, so using deep learning to allow the vehicle to make its own decisions based on what its sensors pick up would be the best way to address any situation.
“Deep learning is the best enabling technology for self-driving cars,” the CEO and co-founder of the company, Sameep Tandon expressed on many occasions that adding that while the sensor input is very important for the vehicles that operate with Drive.ai, what is really needed is a main brain that allows the self-driving car to travel safely and properly by understanding the surrounding environment.