A real box painted black

One of my consistent themes is that cars, which currently account for close to 90% of all person trips, will remain the majority mode in our cities for many decades yet. That will remain true even if governments were to invest in public transport infrastructure on an unprecedented scale.

Policy-makers should accordingly be giving a lot of attention to improving cars and trucks and how they behave in cities. Vehicles should be well-behaved guests in our cities, not our masters. They could do that in a number of ways, but most of them are ignored because they’re politically difficult.

One potential game-changer however is the driverless or autonomous car (see Could driverless cars reshape our major cities?). Theoretically, the driverless car offers many benefits, including significantly delaying the demand for more roads, making travel time more productive, and reducing the environmental impact of private transport.

But they come with potential problems too, like the possibility they might induce more travel and encourage lower densities. The most immediate problem though is implementation, particularly the long transition period when the vehicle fleet would be a mix of autonomous and conventional cars sharing the existing road network.

Driverless cars might represent the technological apotheosis of private transport, but there’s another interim technology that’s considerably simpler. Even so, it could nevertheless have far-reaching effects on the way cars and trucks behave in our cities.

If it were mandatory for every vehicle to be fitted with a ‘black box’, a range of key vehicle behaviours like speed, acceleration and location could be measured and either stored in-vehicle or transmitted to a central repository. (1)

Data could be gathered by sensors connected to the car’s key mechanical and electronic points and by cameras recording the external environment. In turn, vehicles could receive data from local or centralised transmitters.

This isn’t complex stuff in a technological sense. Cars have been operating with on-board computers fed by an array of sensors for decades and an increasing number of new vehicles come with cameras. Transponder and GPS technologies are well developed and some trucks already carry rudimentary black boxes.

At its simplest, the black box would record how well a vehicle conforms to the road rules and provide the driver with real-time feedback. It would be a considerably more powerful however if that information were uploaded to a central database.

It could then be used to enforce compliance with straightforward road rules like speed limits and traffic signals. In the event of a serious incident, data from the vehicle’s cameras and other sensors could be used to investigate the event and possibly provide the basis for police or legal action.

The near-certainty of detection would give drivers a powerful incentive to conform to the law at all times. This in turn should make roads safer for pedestrians, cyclists and other drivers. It should improve the amenity of roadside uses by reducing negatives such as (measurable) vehicle noise.

It could also empower vulnerable road users. For example, a cyclist who feels her safety’s been compromised or she’s been intimidated by a driver could lodge a complaint for investigation. Authorities could review the circumstances based on information uploaded from the driver’s vehicle.

The system could also provide a centralised data base of real-time information on every vehicle ennabling traffic authorities to better manage traffic flows on the fly. That’s not novel either – data from in-car GPS devices is already “crowd sourced” and collected by manufacturers. A mature black box system should also lower policing costs significantly.

There could be other social benefits too. If the way motorists drive is heavily constrained they’d necessarily drive in less fuel-intensive ways and see less value in powerful vehicles. They might not be prepared to drive as far either, leading to more compact land use patterns.

The concept isn’t as fanciful as it might sound. This report from a few days ago says the European Commission’s Mobility and Transport Department is proposing to fit all new cars with speed limiters.

Under the proposals new cars would be fitted with cameras that could read road speed limit signs and automatically apply the brakes when this is exceeded… The scheme would work either using satellites, which would communicate limits to cars automatically, or using cameras to read road signs. Drivers can be given a warning of the speed limit, or their speed could be controlled automatically under the new measures.

That sounds a bit technologically clunky, indicating there’s still some way to go yet. It’s a reminder that, while a black box system would not be as difficult to implement as driverless cars, it would nevertheless come with some challenges.

Since data on a vehicle’s travel behaviour would be recorded, it’s certain there’d be concerns about loss of privacy. That would likely be exacerbated if fingerprint or facial recognition technology were mandatory for drivers.

There would need to be an extended phase-in period, as there was with the transition from analogue to digital television. Manufacturers would need to cooperate and incentives (positive and negative) would likely be required to hasten upgrading of the existing fleet.

The politics would be immensely difficult. It’s likely some drivers would oppose the compulsory loss of autonomy and others would object that such close policing of driving would make it “less fun”.

Although much of it could be automated, there’d be a complex and potentially onerous task in processing offences. The road rules and the scale of penalties would need to be reviewed to reflect what would effectively now be close to 100% detection. Speed limiters would probably have to be part of the package.

As with any new technology, the major value would be in finding applications for the data. The full gamut of potential applications probably isn’t known yet, but some of those could be socially undesirable. A key priority would be ensuring both users and authorities have confidence the system is secure against hacking.

While there would be considerable social benefits from mandatory black boxes, many motorists would perceive it as negative. It would be important to find positive ways of selling the idea. A big attraction could be lower insurance premiums. For example, responsible younger drivers who are currently penalised by broad-brush laws may be significantly better off with a black box system.


  1. I’m not suggesting in-vehicle conversations should be recorded as happens with flight recorders in large aircraft.