Advertisement

Remember learning how to first ride a bicycle? Imagine a car teaching itself how to drive with an intricate 3D map, high fidelity lasers and a laundry list of balancing rules. That is exactly what Amar Shah and Alex Kendall, Ph.D.s from Cambridge University, did with a Renault Twizy car, according to New Atlas.

Their company, Wayve, released video of the updated Renault Twizy teaching itself to drive in between lane lines within 20 minutes. With a camera in front and modified ability to autonomously operate the steering, gas and brakes, the car was hooked up to a graphics processing unit that could intelligently analyze the camera data in real time.

A human driver sat in the driver’s seat but handed over full control to the car, not telling it what its intentions were. Every time the car went off the road, the driver stopped it and corrected it. The algorithm “penalized” the car for making mistakes and “rewarded” it based on how far it traveled without human intervention.

"The missing piece of the self-driving puzzle is intelligent algorithms, not more sensors, rules and maps," says Shah, Wayve co-founder and CEO. "Humans have a fascinating ability to perform complex tasks in the real world, because our brains allow us to learn quickly and transfer knowledge across our many experiences. We want to give our vehicles better brains, not more hardware."

Wayve co-founders Alex Kendall and Amar Shah (Credit: Wayve)
Advertisement
Advertisement