Dying to get there

Every day 10 people are killed on Britain’s roads, making driving far more dangerous than plane or train travel. David Fowler looks at the some of the technological solutions being considered to cut this terrible toll.

Confidence in rail travel has been rocked to its foundations by the crashes at Paddington, Hatfield and Potters Bar in the UK. But though these accidents were tragic, the passengers who travel by car rather than train ignore – or don’t know – the fact that the roads are far more dangerous. In 2001 there were 3,443 deaths on the road – on average, nearly as many fatalities as at Potters Bar every day. Yet this toll receives little attention.

However, over the next decade, several factors will combine to improve road safety and there are likely to be dramatic changes in the cars we drive. The government has set a target of reducing road deaths and injuries by 40% by 2010. A new agreement between car makers and the EU, setting targets for reducing pedestrian injuries, is expected to come into force in 2005.

Following the success of the Euro NCAP crash tests, led by a group of consumer bodies, moves are afoot to extend their scope. Transport laboratory TRL has been commissioned to devise test procedures to rate cars on the basis of braking, handling, lighting, ergonomics and visibility.

And electronic aids will become more widespread. Anti-lock braking and traction control are already common, but cruise control will evolve into systems for collision avoidance, while telematics will become more sophisticated, with the ability to recognise when the car is in a speed limit zone, for example. By 2010 steer-by-wire will be on the horizon.

So will your car of 2010 be completely different from today’s? And will much of the driver’s autonomy be usurped by electronics?

Some cars such as Jaguars already have adaptive cruise control, in which a forward-looking radar monitors the distance from the vehicle ahead and intervenes to slow your car if you get too close. But a number of evolutionary steps are needed before the ‘automated highway’ becomes a reality.

First, says Tim Tiernan, manager for driver-awareness systems at component supplier Visteon, will be lane recognition and side object detection. These will monitor a car’s position on a motorway and warn the driver – audibly or by possibly by shaking the steering wheel – if the system thinks the car is in danger of drifting out of lane.

Next, cruise control will gain a stop-and-go function capable of bringing the car to a standstill if traffic in front stops, and moving off again when the queue restarts. This will be introduced for motorways first ‘where the traffic situation is simple’, says Tiernan. A system for urban situations, where the car’s sensors would recognise small objects such as cyclists and pedestrians as well as large vehicles, will take another year of development, says Tiernan. With most manufacturers having by now sourced their component systems for their 2006 models, this means that 2007 is the earliest date this type of system could be introduced to the market.

A number of technical obstacles remain to be overcome. First is the type of sensor. For lane-recognition, camera-based systems work better than radar. For objects approaching from the side, cameras can locate objects very well but are not good at detecting relative velocity. Camera systems do not work well in poor weather. Radar is good for detecting velocity and is not affected by the weather, but not as good at accurately placing an object. Visteon anticipates that a complete collision-avoidance system would incorporate camera-based lane detection, long-range forward-looking radar sensors and short-range radar to detect objects approaching from the side. If also equipped with an enhanced GPS map, a car would sense if it was approaching a curve too fast.

‘But the key task, beyond individual sensors, is to manage the information. That’s a whole new challenge,’ says Tiernan. To prevent the driver being swamped by warnings Visteon is working on a Cognitive Manager System that would sift information from a variety of sensors and prioritise warnings. ‘The idea is to help the driver to be better, not to take control of the car,’ says Tiernan. ‘You’ve got to focus their attention.’

As telematics become more sophisticated they are likely to play an enhanced role in vehicle safety, for example in speed control. Beacons attached to road signs could communicate with a car as it entered a 20mph speed limit zone, for example. This could simply warn the driver; it could ‘tell’ the cruise control to set 20mph as the new maximum speed but leave the driver able to override it; or it could be mandatory.Nigel White, executive engineer for vehicle dynamics at Lotus Engineering, the consultancy arm of Lotus, says: ‘The technology is OK. It just needs someone to say, ‘Yes, it will happen’.’

The government is unlikely to be enthusiastic about the cost of installing beacons to mark speed limits, but the same effect could also be achieved via a satellite navigation system. In future, rather than relying on CD-ROM maps for information as current satnav systems do, third-generation mobile phone services would allow continuously updated maps to be downloaded, including the latest information on speed restrictions, or even the location of bus stops, schools or low bridges.

‘The obstacle to this is not so much technical as logistical: who would own the data and update the map?’ says Mark Fowkes, intelligent transport systems research manager at MIRA. Continuously updated satellite navigation systems could also warn following traffic that an accident had occurred and help to prevent secondary accidents, which can be as serious as the original crash.

Black box car ‘flight recorders’ are another idea for which the technology exists. In the US sensing and diagnostic modules that record engine revs, speed and brake pedal operation in the last five seconds before a crash are used to give car makers information on airbag deployment. But an external push is needed to get more sophisticated data-recording boxes on to the market. Though they would not directly improve safety they would provide valuable information about how accidents occur. MIRA has demonstrated the feasibility of manufacturing a unit containing a gyroscope and yaw meter, capable of recording information about the vehicle’s main controls.

‘The engineering bit is easy,’ says Fowkes. ‘But who will do the analysis? And what would be the motivation for fitting them?’ Car makers are unlikely to see data recorders as a selling point, so they will only be taken up if external pressure is brought to bear via legislation or insurance companies.

Handling could be one of the most contentious areas. Lotus Engineering’s White says: ‘In the last 50 years there has been a huge increase in our knowledge of car handling and stability.’ Typically a 1950s saloon car, under hard cornering, would start to understeer at an early stage. Drivers would find they had to put in more steering input for a given change in direction, and this would act as a warning to slow down.

‘There has been a dramatic increase in capability, moving the performance envelope further towards the limits of the vehicle,’ says White. So a modern car can corner safely at higher speeds while responses to inputs from the driver remain linear, and hence more predictable, across more of the performance range. But the ultimate lateral acceleration – when the car finally loses grip – has not changed much, so when the driver does reach the limits of adhesion the car is likely to be going faster.

Driver aids such as ABS and traction control are ‘homogenising’ behaviour at the limit, says White. ‘They also allow more latitude in changing the stability/ response relationship.’ But there is a trade-off between these two: at the extremes, a train is highly stable but not very responsive, whereas a Formula 1 car is very responsive but not very stable.

In an illustration of how electronics can influence this balance Cranfield University has developed a fly-by-wire Hawk jet that can be programmed to mimic the responses of either a jumbo jet or a Pitts 2B aerobatic biplane.

In 10-15 years cars without active systems such as ABS may no longer be legal. But White foresees cars developing along similar lines to civil and military aircraft. A drive-by-wire saloon, like an airliner such as an Airbus, would have a chassis designed to be stable, but with electronic controls to provide greater response. A sportscar could be like a Eurofighter, designed to be unstable so it is highly manoeuvrable, with the control system providing stability. ‘If you had failsafe architecture for the control system, hardware and software, you could do amazing things with the car,’ says White.

DaimlerChrysler and General Motors have demonstrated driveable research vehicles: the Mercedes Benz F-400 Carving and the GM Hy-wire, with steer-by-wire systems.Looking further ahead, technology originally developed for Lotus Engineering’s electrohydraulic active valve train, which is now being manufactured for one of its clients, could play a role in chassis systems. Miniature versions of AVT actuators could be applied at key points of the suspension system to allow control over wheel camber and backward, forward or lateral movement of the wheel.

‘As our control system hardware develops and becomes smaller we could have direct control over how the wheel moves. We could achieve the chassis engineer’s ultimate objective: total control of the tyre contact patch,’ says White.

The area likely to require the most urgent attention from car makers, however, is pedestrian safety. While cars routinely gain four (out of five) stars in the Euro NCAP crash test, the testing organisation recently criticised manufacturers because most models score only one star for pedestrian safety.

The European car makers’ organisation ACEA has proposed a voluntary agreement with the EU to improve pedestrian safety in two phases, the first applying to new models type-approved after July 2005. Both sides considered this simpler and quicker to introduce than a formal directive. However, the European Parliament, while approving the voluntary agreement in principle, called for the commitment to be enshrined in law. It is hoped this can be finalised by the end of the year but, with car makers already fast approaching the point at which they have to freeze designs for their 2005 models, time is pressing.

Injuries to pedestrians are caused by impact between the car bumper and lower leg (or in 4x4s, the upper leg), the upper leg and bonnet leading edge, and head and bonnet or windscreen. Tests have been devised to simulate these.

The ACEA commitment, in phase one, sets limits for leg/bumper and head/bonnet impacts. Upper leg/bonnet leading edge and head/windscreen impacts are to be monitored only. In phase two requirements are set for upper leg/bonnet impact and the other regulations are tightened. ‘Bull bars’ will be banned and ABS made compulsory on all new vehicles.

Research by MIRA has identified serious conflicts between pedestrian requirements and other aspects of design, says principal engineer for pedestrian protection Gary Brown. The most serious of these is in ‘packaging’. Designers, who want a low bonnet line and short overhangs, fight with engineers over every millimetre of packaging space. The leg impact standards will require extra space behind the bumpers to allow a softer impact, while the head test will require more space between bonnet and engine to limit secondary impact with hard components underneath. This could amount to an extra 20-40mm of overhang at bumper level. A lower ‘stiffener’ to reduce leg injury will also be needed.

There will also be conflicts in aerodynamic balance and engine cooling because of changes in airflow to the radiator. Sport-utility vehicles will be unlikely to meet the leg impact requirements without vastly reducing the bonnet leading-edge height, which will mean they no longer look like SUVs. Headlights are a problem as they have to be mounted rigidly and their weight precludes designing them to detach on impact.

The bonnet line is likely to need to be raised by as much as 55mm in phase one, affecting sight lines for some occupants. In phase two, in many cars the driving position will have to be raised, meaning that the roof will also have to be higher, affecting all aspects of body design from crashworthiness and torsional stiffness to aerodynamics and fuel economy.

As a result, car manufacturers are showing interest in active safety systems. Pyrotechnics could be used to raise the bonnet automatically in the split second before impact to provide more cushioning space, provided a sensing system could be developed that would reliably detect a pedestrian. This would allow current styling trends to be preserved. A three-year European research programme, SAVE-U, to develop suitable sensors, started earlier this year but its results will not be known until phase one is about to come into effect.

In the end, do all these developments mean driving will be less enjoyable? Lotus Engineering’s White believes not. Though electronics may be taking more control over the car in panic situations, ‘our role is to make it fun in the linear range’, he says. White believes Lotus’s research into vehicle dynamics puts it in a good position to identify what makes one car more fun to drive than another: ‘We’re looking at ways to inject ‘Elise-ness’ – despite legislation.’

Sidebar: One of the riskiest forms of travel

More than one in four of all accidental deaths in the UK occur on our roads. Provisional figures for 2001 show there were 3,443 road deaths and 37,094 serious injuries. In 2000, the latest year for which a detailed breakdown is available, the death rate for car passengers was 2.8 per billion passenger kilometres. This compares with 0.02 fatalities for air travel, 0.3 for bus and coach, 0.4 for rail, and 0.8 for water transport.

The situation for pedestrians, motorcyclists and cyclists is worse, with 48 pedestrian fatalities, 30 for pedal cycles and 130 for motorbikes, mopeds and scooters.Nearly twice as many car drivers as passengers were killed or seriously injured: 12,695 to 7,024. Of the drivers, 3,219 were aged below 25.

Of 3,039 road deaths in 2000, 2,252 were male and 787 female.

Over 10 times as many deaths occurred on A-roads as on motorways: 1,986 to 189.The UK, though, compares well with other countries, with 7.4 road deaths per billion motor vehicle kilometres compared with 14.9 in Austria, 12.2 in Germany, 9.6 in the US, and 119.8 in Turkey.