Project Success Stories - Smart cars for chaotic roads
Day by day, roads seem to become more chaotic. Drivers are no longer humanly capable of seeing all that needs to be observed to reduce the risk of a crash. In fact, with cars, we have out-paced our biology ― the human nervous system evolved to handle, at most, the sensory changes that might happen while running down a steep hill. We drive cars more than five times as fast. We might have coffee in one hand and the mobile phone in the other and, at the same time, we are glancing at the global positioning system (GPS) screen. Many drivers are guilty of paying less than 100 % of their attention to the road. In spite of the tremendous advances made in crash protection technologies, tens of thousands of people die each year on Europe's roads and more than 1.5 million are injured. After years of research, crash prevention is beginning to take centre stage in the battle to reduce the harm caused by car crashes. For the 'Relative positioning for collision avoidance systems' (Reposit) project team, this did not necessarily mean creating new safety gadgets. Instead, it meant bringing the principles of computer science to bear on the different systems already making their way into vehicles. The team started with the assumption that, in a relatively short time, vehicles and roads will be equipped with sensors and communications technologies. There are already projects in Europe and the United States that are testing roadside sensors which can wirelessly relay information about accidents, traffic congestion and weather to an interface on board cars. The 'Collision warning system' (CWS), the brainchild of the Reposit researchers, uses new systems which are starting to become standard automotive features ― vehicle-to-vehicle (V2V) communications systems. Built on low-cost technology that combines GPS with a wireless networking technology similar to household Wi-Fi, V2V systems can make driving safer by alerting motorists of potential collisions. Imagine, for example, a child running across the street unexpectedly just when a car approaches at speed. It's too late to slam on the breaks. The driver cannot swerve either, because there is an oncoming car in the next lane. By helping both cars to coordinate their movements together, CWS could take the edge off such a situation in the near future. Cars recognise potentially dangerous situations ? High-performance GPS, which can locate a car within a metre or so, is used to determine the relative position of neighbouring cars. The vehicle-to-vehicle communication link provides further information about their speed, indicates cars moving at slow speed or even warns if a car ahead has braked hard. Using information from these two systems, CWS then extrapolates where they will be in a few seconds time. If the calculations predict an imminent collision, it warns the driver. The Reposit team tested a fully functional prototype of the new system in a laboratory simulator and the results were encouraging. 'We've got predictions about one to three seconds ahead of collision […] but anything from 2 seconds up gives drivers time to react,' reveals Jose Ignacio Herrero Zarzosa, coordinator of the Reposit project. To validate the accuracy of the system under different driving conditions, more than 70 simulation tests were carried out. In rural and suburban roads, where vehicle speeds are typically higher ― above 50 km/h ― CWS has proven to work better. Even in urban settings where there are many obstacles potentially impeding V2V and GPS signals, reasonable warning times of at least two seconds could be obtained. Using its available sensors, the car can be more or less aware of what's going on around it. The next question is, what information is important to the driver and how does this information get relayed? Thanks to the numerous driving simulations, Mr Zarzosa and his team were able to determine how all the data gathered by a high-tech car can be prioritised in a way that is most helpful to the driver. The Reposit 'on-board unit' provides different visual and acoustic cues, depending on the risk of an imminent collision. Similarly to a street light, a 'green' warning sign implies that there is no risk of a collision. The purpose of a 'yellow' or an 'orange' warning sign is to increase the alertness of the driver about the danger of a potential impact against a neighbouring car. When the danger level is raised to 'red', all neighbouring cars are informed to allow drivers more time to react. While not as smart as the human brain, the new system never becomes tired or distracted. It will be looking in every direction while the driver looks in one place at a time ― a potentially life-saving advantage. Where to from here? The new technology works in the lab. CWS is not the only collaborative driving system designed to prevent accidents and reduce traffic congestion, but it does have a significant advantage over other similar systems. CWS needs no further roadside infrastructure apart from the on-board unit. Have car manufacturers shown interest? The Reposit project was also the first serious effort to use GPS information for crash prevention. The automotive industry finds the project's research work interesting but there is still some convincing to do on the commercial side, suggests Mr Zarzosa. 'The car industry is very price sensitive,' he says. 'Right now, there is no standard for integrating new functions into an existing car system. Every manufacturer uses different system integration methods,' he adds. 'This significantly pushes up the cost of third-party technologies like CWS.' The European Commission reports that it is working hard on easing the burden of integrating new features through developing common standards. Even so, the last challenge for the Reposit team remains affordability. To allow faster market penetration, they left the door open to a version of CWS that could be installed on the roadside. This could be part of the solution, as car accidents may not involve only cars with an installed collision warning system, but also bikes and pedestrians crossing the road. The Reposit team also turned their attention to the original equipment manufacturer (OEM) GPS boards, which are now appearing on the market and offer an attractive trade-off between performance and price. Their new goal is to reduce CWS to a software module that could be added to the on-board GPS system. Before long, the pace of introducing crash prevention systems is expected to accelerate, as 'a cooperative approach' is the only way forward, according to Mr Zarzosa. In the near future, when warnings will be issued but there's not enough time to react, crash prevention systems may also operate the brakes and steering wheel to keep the car occupants and pedestrians out of harm's way. The Reposit project received research funding under the EU's Sixth Framework Programme.