\n
![](\"https:\/\/techcrunch.com\/wp-content\/uploads\/2019\/04\/wayve.gif\")
<\/p>\n [ad_1]
\n<\/p>\n
Projects like Google\u2019s Waymo, <\/a> Uber, Cruise and Aurora are developing autonomous vehicles by throwing engineers at the problem, basing most of their platforms on rule-based systems that try to pre-empt and deal with every edge case, whilst also peppering the cars with more sensors to capture more data. This can work in relatively controlled environments but has the drawback of not being able to flexibly adapt in real-time to fast-changing situations.<\/p>\n Despite all this investment and many years of development, no one has yet been able to launch a commercial autonomous car service. It\u2019s just very hard to hand-engineer. What\u2019s required is not more eyes but better coordination. The simple answer \u2014 as it is to almost everything these days \u2014 would be to throw AI at the problem, and that\u2019s what many startups, which lack the engineering and hardware muscle of the big players, are trying to do.<\/p>\n We reported<\/a> on UK start-up Wayve<\/a> last year when it announced its existence, but they had nothing to show for their claims.<\/p>\n Now they say they do, and the results are not only fascinating but might also be genuinely innovative.<\/p>\n In fact, they are claiming a \u201cworld first\u201d in demonstrating that a car working on their machine-learning platform can drive on roads it\u2019s never seen before during training, and without an HD map of its environment. Other systems, like Waymo\u2019s, rely on maps and rules to drive. Theirs, says Wayve, does not.<\/p>\n Additionally, it\u2019s also revealed that it\u2019s been testing its platform on the Jaguar I-PACE SUV, which won the 2019 European Car of the Year. Interestingly, this is also a car which has been used by Waymo in some tests.<\/p>\n Alex Kendall<\/a>, Co-Founder & CTO tells me: \u201cOur cars learn to drive from data with machine learning. Every time a safety driver intervenes and takes over, the car learns to drive better. We don\u2019t tell the car how to drive, rather it learns to drive from experience, example and feedback, just like a human. This is more safe and scalable than any other approach today.\u201d<\/p>\n Emerging out of research from the University of Cambridge, <\/a> Wayve has already undertaken extensive testing on public UK roads.<\/p>\n Kendall says: \u201cThe traditional approach used by all our competitors relies on HD-maps, expensive sensor suites and hand-coded rules that tell the car how to drive. We have built a system that learns end-to-end with machine learning. It is the first in the world to drive on urban roads it has never been on before. It uses compute\/sensors which cost less than 10% of competitors.\u201d<\/p>\n Bold claims. But a video today revealed<\/a> on their site shows their system driving on public roads in Cambridge, UK, driving on roads it has never been on before using a sat-nav route map and basic cameras.<\/p>\n