The UK took further steps recently towards defining the future rules and driver obligations for self-driving vehicles on the roads. The Department for Transport (DfT) and the Centre for Connected and Autonomous Vehicles (CCAV) published changes to the Highway Code this week that are going to help ensure road safety as the first wave of self-driving vehicles are due to enter service.
According to the DfT’s definition, the first self-driving cars could be “ready for use” in Britain later this year, to support early deployment and understanding of the technology. A full regulatory framework including the insurance industry’s roles and responsibilities, is expected to be implemented in 2025.
This new set of measures follow a public consultation launched by the government, which found the majority of respondents were broadly supportive of the proposed changes to the Highway Code to clarify a driver’s responsibilities when a vehicle is in self-driving mode.
The introduction of the technology is likely to begin with vehicles travelling at slow speeds in closed areas on private campuses, or on motorways where clear lane marking, road layout and infrastructure is more consistent than on other roads. This is typically in congested traffic and certain other defined situations where a driver can legally switch to ‘hands off’ and watch in-vehicle TV or engage in other activities. However there is an exception for mobile phone use, which will still be prohibited. The driver must remain able to take back control of the vehicle and it is this transition that is likely to be the topic of much debate and consideration.
However, the reactions to the government proposals in the media and social media show the difficult road ahead for public education and the role of insurance companies, the technology and automotive industries, in reassuring road users.
Transport expert and cycling advocate Christian Wolmar, who sits on the board of the London Cycling Campaign, commented that the government had failed to make the essential distinction between fully self-driving cars and Level 3 vehicle autonomy as it exists on the road today where the driver’s full attention is still required at all times.
“There is no halfway point where you can watch films [in a vehicle] but still be ready to take over. This will kill people,” he said.
“Self-driving cars: Motorists will not be liable for crashes and can watch TV behind the wheel,” said a Sky News headline. “Self-driving car users will not be responsible for crashes,” said the Daily Express. “Self-driving cars are coming and they’ll flash you an emoji,” said the Times.
It’s a direction of travel with new technology and public perception that would be served better by a more responsible and measured approach from the mainstream media. It’s going to increase the responsibility of the insurance provider to understand and identify the fitment of technologies on a specific vehicle and understanding of the system’s capability and related history of insurance claims.
New solutions such as LexisNexis® Vehicle Build, which returns all ADAS features, their behaviour and purpose, identified on the vehicle and appended to the Vehicle Identification Number (VIN), are going to be key to understanding how self-driving vehicles perform in the context of risk in real-world driving conditions.
And, it’s not just how the data is collected in relation to the profile of a particular vehicle, but how it will be standardised, validated, analysed, and ultimately shared into insurance processes that still needs serious cross-industry effort.
It is surprising that the government didn’t make much reference to the road safety challenges and the nitty gritty regulatory challenges that still need to be refined in detail. This seems to betray some lack of understanding of the skills that already exist in the insurance market to determine risk, based on driver behaviour or vehicle behaviour. It also underestimates the expertise needed to bring insights from the vast amounts of data to come from ADAS (Advanced Driver Assistance Systems) and in future the dynamic, real-time data from the connected car.
The fact is we need as much investment in data science skills as data technology. One cannot work without the other and it’s a massive scale of investment that is practically impossible for a single insurance provider to achieve by themselves.
The success of being able to take account of ADAS fitment in the context of risk relies on the growing volumes of data that can be obtained and understood from those vehicles. At LexisNexis Risk Solutions we’re continuing to work closely with the automotive OEMs and other data partners to ensure that the data is in a usable format for insurance.
By gaining access to the detailed data on car systems, insurance providers will be able to understand how risk changes across the different levels of autonomy and develop insurance products accordingly.
In response to the government’s announcement, Steve Gooding, Director of the RAC Foundation, said: “These latest additions will help us all understand what we must and must not do as we move forward to an environment where cars drive themselves.….The final part of the jigsaw is to ensure these amendments are widely communicated to, and understood by, vehicle owners. Vehicle manufacturers and sellers will have a vital role to play in ensuring their customers fully appreciate the capabilities of the cars they buy and the rules that govern them.”
Taking the next steps to vehicle autonomy with ALKS
The government announced in April last year that vehicles fitted with automated lane keeping system (ALKS) technology could be the first example of self-driving technology. Existing ADAS technology available on the market is ‘assistive’ or ‘passive’ rather than ‘active’, meaning drivers must currently always remain in control and responsible. Designed for use on a motorway in slow traffic, ALKS enables a vehicle to drive itself in a single lane, up to 37 mph, while maintaining the ability to return control easily and safely to the driver when required.
It’s important to state that there are many, many societal benefits to come from self-driving vehicles in the long run, not least of which is the likely lower cost of motor insurance.
The technology can improve and level up transport options (eventually for non-drivers), help to reconnect rural communities, ease congestion, whilst cutting emissions and reducing collisions caused by human error, which is a contributory factor in 88% of all recorded road collisions.
Thousands of new jobs and new data challenges
The government estimates the development of self-driving vehicles could create around 38,000 new, high-skilled jobs that would be worth over £40 billion by 2035. But when it comes to assessing the real risk on the road, from the data that we hold at LexisNexis® on ADAS and the impact on pure premium losses, it is a very complex picture of risk to date.
There are variations in the loss cost impact seen when looking across different technology types and clusters, different policy coverage types (for example third party versus own damage). There are variations across the different markets where we operate (for example UK, US and Spain) and when comparing claims frequency and claims severity.
It is an ongoing exploration with data that is continuing to deliver better and better underwriting results as we continue to grow our data sample. The immediate challenge for insurance providers is that these interim measures announced from the government, which are set to come into effect in the summer, include making insurance companies, responsible for liabilities in certain circumstances when a vehicle is in self-driving mode, rather than the driver.
The insurance industry has the benefit of years of experience and data for analysing road traffic accidents and liabilities. In practice there are still grey areas to be studied, related to possible product liability related to vehicle technology, analysing the seconds and milliseconds transitioning from self-driving to driven mode, and complex insurance claims involving the interactions of self-driving car technology and a human-driven car. But in practice vehicle-derived data is already starting to help insurance providers with their pricing and underwriting decisions.
Follow the link to the LexisNexis Risk Solutions website to find out more about how we support insurance providers.