Truck stop

7 min read

Heavy goods vehicles are involved in 10 per cent of all fatal road accidents in the UK. Could an ambitious European project to develop sensing and control systems reduce the carnage?

On Wednesday 25 October, a five-lorry pile-up on the A66 in Cumbria left one man dead and four others injured. In November, dozens of children in Yorkshire had a narrow escape when a cement-mixer lorry crashed through the railings of their school playground. On the same day, a woman was killed on the M5 when a heavy goods vehicle (HGV) crossed the central reservation and crashed into her car. The list goes on.

Owing to bad weather, driver error or mechanical failure, there are many crashes involving HGVs. And while accidents caused by cars are certainly more numerous, crashes caused by trucks are responsible for a disproportionately high number of fatalities.

Department of Transportstatistics show that in 2005, rigid and articulated HGVs accounted for 1.34 per cent of licensed vehicles in the UK but were involved in 3.31 per cent of all accidents and 10 per cent of all fatal accidents.

Truck safety clearly needs to be improved if there is any hope of achieving the EU’s goal of halving the number of road deaths by 2010. In 2000 the number of fatalities topped 52,000 in the 25 states that now make up the EU.

Brussels is banking on new technology to tackle the problem. Under the PReVENT banner (short for ‘PReVENTive and Active Safety Applications’), a number of EU-supported projects that will run until 2008 are bringing together researchers from major commercial vehicle makers such as Volvo and DaimlerChrysler, as well as a raft of components suppliers.

Together the participants aim to develop integrated sensing and control systems for a number of purposes: to prevent trucks from straying from their lanes, to stop them from colliding with vehicles in front, to warn drivers of difficult road conditions, and even to help drivers take corrective action. Some systems the researchers are working on are very ambitious and may not be seen in production vehicles for many years, if ever. But some crucial improvements to basic steering and braking functions that might save lives could be based on existing systems and may appear on Europe’s roads by 2010.

The state of the art today is exemplified by DaimlerChrysler’sMercedes-Benz Actros ‘Safety Truck’. An optional safety package available for the Actros includes proximity control, lane departure warning and stability control.

The proximity control is similar to the radar-based adaptive cruise control systems increasingly found in high-specification cars. When the system detects a slower-moving vehicle ahead it automatically applies as much as 30 per cent of the truck’s braking power to restore the correct following distance set by the driver. If the gap widens or the truck moves to a clear lane, the system automatically accelerates the truck back to its pre-set speed.

Like the electronic stability programmes (ESPs) available in many passenger cars, Mercedes’ stability control system for trucks uses lateral acceleration sensors to detect the oversteer or understeer that can lead to skidding or rollover. Then, it keeps the truck stable and on course by controlling the throttle and braking individual wheels — including those on the trailer. The system relies on existing antilock braking technology to operate braking for individual wheels.

A camera-based system available since 2002 on the Actros also monitors the vehicle’s position relative to lane markings. If it detects that the truck is straying from its lane, the driver is alerted by a noise similar to that produced by a corrugated surface on the relevant side. According to Mercedes, this will cause the driver naturally to steer to the opposite side.

Road users who fear rear-end collisions will take comfort from Mercedes’ most recent innovation. Since September, Actros trucks have been available with an active brake assist system that, using the proximity control radar, can automatically initiate emergency braking if a collision is unavoidable and the driver doesn’t take action.

Active brake assist uses three radar sensors to monitor the road between seven and 150m ahead. If a collision with a vehicle becomes unavoidable, the system first warns the driver with an illuminated red triangle on the dashboard and an alarm. If the situation gets worse then partial braking, and if necessary emergency braking, is automatic.

DaimlerChrysler technology spokesman Claws Tohsche said that, at the very least, active brake assist offers the great benefit of reducing the impact of unavoidable collisions. It costs only €300 to €400 (£200 to £270) more than the basic safety package because it only requires a reprogramming of the intelligent cruise control system.

But taking the next step in truck safety will be more difficult, according to DaimlerChrysler. When introducing the Safety Truck, board member Hartmut Marwitz noted that fatalities involving trucks in Europe fell by a quarter between 1992 and 2004 despite increasing traffic. But he added that traffic is forecast to carry on rising — by around three-quarters by 2030. ‘So it will be extremely difficult to achieve the aim of [further] reducing accidents,’ he said.

The PReVENT programme includes projects to improve control of speed, following distance, lane keeping, blind spot visibility and collision avoidance and mitigation. To improve significantly on the Actros of today, each system will have to sense and analyse the road environment with much greater accuracy than is currently possible.

To improve collision avoidance, for example, systems will have to differentiate very reliably between a vehicle stopped in the lane ahead and one parked at the roadside. And to improve lane-keeping assistance they will have to accurately assess difficult situations — for example when there are no lane markings or when the markings have been temporarily altered for roadworks.

The leap will be large from concept and testing to a production vehicle that can safely and reliably intervene to assist or even override the driver.

For its part, DaimlerChrysler has marked out its research path with a shopping list of 50 developments. These range from reactive steering to compensate for crosswinds to 3600 visibility cameras and electronic stability programmes enhanced by active steering control and curve prediction.

But as Tohsche readily admitted, only a few of these concepts could be road-ready soon. DaimlerChrysler’s top priority, he said, is to adapt its active brake assist system for the stop-start conditions of a slow-moving traffic jam. This poses a big challenge for sensors, Tohsche added. The first step will be to create systems to sense snow and ice on the road, and reliably differentiate between standing vehicles and fixed objects, which is particularly difficult.

Many of the PReVENT technologies have been brought together by Volvo, which showed them in its Integrated Safety Truck demonstration at the Intelligent Transportation Systems World Congress in London in October. Malte Ahrholdt, a systems engineer at Volvo Technology working on the PReVENT project, agreed that the main challenge for advanced active braking systems will be to perceive the environment reliably enough to decide when a collision really is unavoidable.

Misinterpretation would be a critical failure if it caused automatic braking when there was no collision situation, Ahrholdt said. What makes the problem so challenging is the fact that while direct and accurate measurements can be made of a vehicle’s own condition — speed, direction and driver’s steering input, for example — measurements made by radar or other sensors of the surrounding environment are always indirect and comparatively unreliable.

Therefore, the trick will be to eliminate ‘ghost objects’ by bringing together data from a number of sensors, according to Ahrholdt.

Volvo’s Integrated Safety Truck uses a laser scanner to anticipate possible collisions. The system’s automatic braking can’t bring a truck to a complete stop, but can reduce the impact of a collision. Emergency braking to a full stop is possible but not yet reliable; Ahrholdt said that more experience with active brake assistance systems is needed to safely introduce that level of automatic intervention. Complicating the situation is the fact that, given warning, a driver may take evasive action before or even during automatic emergency braking. To avoid causing accidents a system would have to reduce or cease its intervention in such cases.

Braking systems based on these principles are already appearing in cars and are entering series production. However, Ahrholdt stressed that aggressive braking assistance is much more difficult to achieve in trucks because the dynamics vary more widely, and depend on the vehicles’ loads and trailer configurations.

One of Ahrholdt’s colleagues, Achim Beutner, is working on steering assistance with an emphasis on lane keeping. Here, cameras and digital maps monitor the Integrated Safety Truck’s position in the lane, and its electric power steering system provides a slight torque to the steering wheel to help guide the driver back to a correct position when necessary.

Another technique is to generate a smooth vibration in the steering wheel when correction is needed. Beutner prefers this torque or vibration to Mercedes’ audible warning — a sound which may annoy some drivers. The torque is a gentle assistance that helps make driving more comfortable, he said. Beutner emphasised that the system is not strong enough to override the driver, who is always in control. Drivers who have tested the system like it, he added.

Volvo’s concept is technically not far from being practical for series production. However, this step won’t be economically feasible until electric power steering systems are widely available on trucks, which have yet to move beyond traditional hydraulic systems.

One of the most promising systems in the Volvo Integrated Safety Truck is its Start-Inhibit function that prevents the truck from accelerating if its stereo cameras and radar detect an obstacle in the front blind spot. Gustav Markkula, who oversees Volvo Technology research into driver-vehicle interface, said Start-Inhibit can intervene without adding any risk, as holding a vehicle stationary would be unlikely to cause an accident.

All these pioneering systems could be improved by data from satellite navigation systems. Vincent Blervaque of Brussels-based intelligent transport systems researcher Ertico said that by enhancing existing digital maps with data about lane markings, speed limits, height and width restrictions, recommended truck routes, and so on, more information could be fed into safety systems to help them better predict hazards by allowing them to ‘see’ further than their sensors can detect. The addition of information about slopes may also enhance stability control and improve fuel economy, especially for hybrid vehicles.

Some of these applications could be available within five years, Blervaque predicted, because some advanced cruise control systems already make use of such navigation data.

Of course, drivers are the most important component of any vehicle safety system. So it’s not surprising that monitoring their alertness and performance is another hot research topic. Here, Volvo and Mercedes take different approaches. Markkula is enthusiastic about camera-based systems that measure blink rates or detect a nodding head. These systems are already ‘quite advanced’ and have achieved good results in tests, though they are currently too expensive for widespread use, Markkula said.

However, DaimlerChrysler’s Tohsche questioned the reliability of this approach. He said Mercedes prefers technology to monitor steering and throttle movements for signs of tiredness or distraction.

But whichever hi-tech systems are finally offered to the public, the life-saving potential will be undermined if drivers continue to overlook basic safety routines. Tohsche put it bluntly: ‘A lot of them still need to learn to wear their seat belts.’