Wayve, a British startup, announced on Wednesday that it would handle massive volumes of data using supercomputer infrastructure built for it by its investor Microsoft as it develops machine learning-based models for self-driving cars.
Instead of depending on precise digital maps and code to inform vehicles how to operate, Wayve’s technology depends on machine learning using camera sensors mounted on the outside of the vehicle, where the system learns from traffic patterns and the behaviour of other drivers.
“Microsoft is providing supercomputing muscle,” Wayve Chief Executive Alex Kendall said. “What we’re looking to do goes beyond the bounds of what’s possible for commercial cloud offerings today.”
Kendall claims that Microsoft will be able to process the terabyte of data – 1 trillion bytes, or equivalent to around an hour of consumer video – that Wayve’s cars generate every minute.
This will aid the business as it expands its self-driving technology in preparation for testing on last-mile delivery vehicles with UK online grocery retailer Ocado and supermarket chain Asda.
The trials for grocery delivery will begin this year, with a human safety operator on board.
“We see this as being a commercial fleet offering,” he said. “That’s how we think autonomy is first going to come to market.”