After undertaking a year-long investigation with Ford and four other mobility specialists on how to build self-driving systems that integrate with London’s existing transport infrastructure, Addison Lee today is announcing the next step in its autonomous strategy.
The on-demand ride company — which competes with black cabs, Uber and other car services — announced a deal with self-driving startup Oxbotica to develop autonomous vehicles, with the aim of getting them in service in London by 2021.
No financial details are being revealed about the deal. Addison Lee CEO Andy Boland, in an interview, described it as “purely commercial.” Currently Addison Lee is “unfashionably profitable,” he added, and so it is working on its current self-driving efforts off its own balance sheet. It is also wholly owned by PE firm the Carlyle Group, so it’s likely that this will help with future funding, although Boland did not rule out that when the company gets closer to a commercial launch, that it might need to look for funding to do this.
Meanwhile, Oxbotica — a spin-out from Oxford University — has raised around $18 million to date, with backers including Oxford, Innovate UK, the Ministry of Defence, the IP Group, insurers Axa XL and others.
The deal potentially sets up an interesting new avenue in how we might see autonomous cars being built, rolled out and operated.
Today, a number of transportation-on-demand companies that have roots in the tech world (Uber, Didi and Yandex Taxi are three examples) are trying to build their own self-driving tech. Alongside them, there are also a plethora of car makers who are also intent on building and running that experience.
Now, Addison Lee and Oxbotica are essentially presenting a third option: a system not built by the car service-operator, and not by a car maker, but by a third-party tech company that overlays the tech on top of whatever vehicle the service provider chooses to have.
Oxbotica was the first company to get a green light to start any kind of self-driving car test on a UK road, when it tested its equipment on a modified Renault driving five miles per hour in the town of Milton Keynes in 2016.
That early start was one of the reasons why Addison Lee decided to go with Oxbotica for its own efforts.
“The way they have built their technology and how they are already provisioning it has been impressive,” said Boland. “They have the most tangible capability that we could go with, and we felt that the way they were thinking about it, the business model isn’t just for ride share or car share, it’s across a range of other industry applications, and we like that, too. It’s now getting serious, and real-life-scale operators like Addison Lee are looking to bring this to market.” That would include shuttle services but potentially other kinds of commercial transportation (note that Oxbotica counts a commercial insurer as a backer).
“This represents a huge leap towards bringing autonomous vehicles into mainstream use on the streets of London, and eventually in cities across the United Kingdom and beyond,” said Graeme Smith, CEO of Oxbotica, in a statement. “Our partnership with Addison Lee Group represents another milestone for the commercial deployment of our integrated autonomous vehicle and fleet management software systems in complex urban transport conditions. Together, we are taking a major step in delivering the future of mobility.”
The two will start first with a comprehensive 3D mapping sweep before moving on to other aspects of building the service to prepare a autonomous vehicle service for trials.
For now Addison Lee has not named a vehicle partner, and Boland said that this was intentional.
The initial mapping exercise will be on the company’s existing fleet of vehicles working on autonomous research around London today – those will be the Fords from last year’s deal, he said. “But in terms of future service provision in 2021, that decision is yet to be made. Oxbotica is agnostic to manufacturers, and that was also interesting to us at the moment.”
Although Addison Lee competes with the likes of Uber — which had to appeal for a provisional 15-month right to operate in London after initially losing its license — it is different in that it owns its own fleet of vehicles. That has given the company more of an incentive to try to develop its own technology that it might use across whatever vehicles it chooses to use in its fleet. It’s not clear how this will work alongside the fact that many car makers are also working on their own autonomous technology and subsequent strategy to “own” the self-driving experience — at some point they might even directly conflict — but for now following this route could be Addison Lee’s strongest bet for continuing to keep a close service for its drivers and its fleet, developing it in the way it feels is best for its own business.
“It’s about having a bit more control between us,” Boland said of the deal with Oxbotica, ” and creating and building those services together without the dependencies without the changing priorities of an OEM. So that we can do what we need to do.”
The self-driving car market is expected to be worth some £28 billion in the UK alone by 2035 , and car services — versus private ownership — are shaping up to be a large part of how it might play out. Startups like FiveAI are also focusing on building self-driving car systems, specifically so that they can run in their own fleets of vehicles and in a service run by FiveAI: its reasoning is that autonomous vehicles will be too cost prohibitive to own for the majority of people.
As car ownership is predicted to decline as cars become ever-more sophisticated and expensive, car services are already on the rise and are predicted to grow some 21 per cent by 2030 .