Just a few months ago, Tesla’s founder, Elon Musk, compared the company’s Autopilot AI system to Google’s search engine, acknowledging that they believe their system is superior to other driver-assistance systems in the industry.
The company’s electric cars are equipped to stop automatically at lights and stop signs, and can control acceleration and steering. The drivers’ actions assist in training the computer over time and communicate data back to Tesla’s engineers to help with technology improvements and future software updates. This process is very similar to Google’s search engine technology which learns based on what users search for and click on.
But what makes Tesla’s unique driver-assistance software a cut above the rest? Or what they consider superior?
The History & Future of Autonomous Driving
Aspects of autonomous driving have actually been present for years. Features that we all have utilized like cruise control, lane change sensors, ABS braking, and even airbags were just mere baby steps towards the advanced driver-assistance system that Tesla has now developed.
Tesla’s system functionalities are currently classified between level 2, “Partial Automation” and level 3, “Conditional Automation” under the U.S. Department of Transportation’s five levels of vehicle automation. This means the vehicle has combined automated functions and can monitor the environment but the driver must be ready to take control of the vehicle at all times.
Musk previously claimed that 2019 would be the year for Tesla to release its fully autonomous driving system based on the original Autopilot software. He has since changed the timeline to the end of 2020.
I feel extremely confident that it will be possible to do a drive from your home to your office most of the time with no interventions by the end of the year,” he said. “We can almost do this already with the leading-edge [Autopilot] driving the car.
Reaching level 5, “Full Automation” is no easy feat. Think about it – how can you ensure that a vehicle based fully on computer functionality can think, react and make safe decisions when on the road?
(A view of what Tesla Autopilot’s neural network sees on the road. Source: Carscoops)
Tesla has taken a novel approach in this realm – their neural network. Their fleet of almost 1 million vehicles “take video to capture road layout, static infrastructure and 3D objects.” Each vehicle records data from its eight cameras, 12 ultrasonic sensors, an inertial measurement unit, drivers’ actions and GPS.
So, as a Tesla vehicle owner, you’re helping the company to research and improve their capabilities to move towards more advanced self-driving capabilities.
Tesla currently claims that all of their vehicles are already equipped with the hardware necessary for full self-driving, and that they only need software updates to reach full autonomy.
Only time will tell if 2020 will be the year that Tesla reaches full autonomy and possibly changes the landscape of driving for the future. I for one, am very curious if and when this technology will be available and approved for public-use.