Autonomous driving could be described as active safety in its most advanced form. It will be accomplished when and if technology allows vehicles to be completely self-operating, independent of any input at all from the driver. In reality this requires vehicles to have driving capabilities comparable to those of an experienced human driver with the addition of being able to handle critical traffic situations reliably and that the systems are meticulously tested. There is a clear trend among prototype developers to create technology for self-driving vehicles, but an enormous amount of research and development is still required if the systems are to be sufficiently safe and reliable throughout a vehicle´s lifetime.
Intelligent communications systems can be used to realize better active safety systems. The basis of these new systems is that they will make it possible both for vehicles to communicate with other vehicles and for vehicles and infrastructure to communicate with each other. (V2V and V2I). Precise localisation with international satellite navigation systems such as the US GPS, Russian GLONASS and the forthcoming European Galileo in combination with communication capabilities can enable interconnected protective systems. Highly accurate positioning of test vehicles (initially via GPS and differential GPS) is required as reference on test-tracks in order to test and develop this technology.
Road users without vehicles such as pedestrians, cyclists and light moped riders are the most vulnerable to traffic accidents. As vulnerable road users account for a large proportion of global road traffic deaths, developing safety systems to increase their protection is paramount. This is a complex area that relates to everything from integrated traffic management systems to how we can develop and improve warning systems that alert these road users. Hybrid buses, for example, are very quiet compared with normal car engines and can therefore be difficult to detect, not least for the blind. As people in traffic are trained to use their ears to detect danger, it might sometimes be necessary to create artificial noise for silent vehicles.
Test targets for active safety need to be equipped with camera and vehicle to x-communication since they will be part of an interconnected system. For example, it must be possible to mount these targets on flat carriers and to link and connect them with GPS and localisation systems. They can thus be programmed for predefined trajectories. In turn it must be possible to equip the carriers with cameras in order to analyse nearby test collisions. Several different embodiments of test targets will be required, target vehicles with metallic surfaces of different vehicle sizes for example, and test targets for bicycles, motorcycles and pedestrians (children and adults).
Together with researchers, AstaZero aims to build a catalogue of methods for how to provoke false positives in order to test the robustness of active safety systems, confirming whether the algorithm is safe and reliable enough. Here methodology needs to be suitable for the actual test performed as, for example, camera-based and radar-based systems have different characteristics of false positives.
Advanced tests require careful pre-planning in order to ascertain the correct scenarios. Although some tests at the AstaZero test site can be standardised, most tests probably need to be conducted first in an AstaZero 3D virtual environment. It is only then that the researchers will know which scenarios are most critical, and will then be able to focus more specifically on them. This is also a good way to guarantee the quality of the test, ensuring that the time spent at the test site is not wasted by selecting the wrong test scenarios or even technical problems, which can be ruled out with proper test preparation based on simulations.
Human factors such as distractions or tiredness are the main reason behind most traffic accidents. Modern cars will therefore increasingly have electric controls, including warnings that alert the driver to the risk of an accident situation. Advanced systems are under development that detect whether the driver is dozing off, stressed or under the influence of alcohol. When developing these systems, account must be taken of the fact that people react differently to specific situations, and extensive research and testing is therefore necessary to develop driver models that produce reliable systems.
The introduction of electric vehicles has brought with it some new risks that have to be tested, not least for new active safety functions. Apart from the more common problem of how to store batteries in the car with respect to centre of gravity and weight, there are also several aspects related to how to heat vehicles sustainably. Furthermore, safety also has to be guaranteed for electrification in an accident situation, for instance due to a voltage drop or where there is an electric shock or a fire. It is important to test whether, and the extent to which, the safety systems still work if they are involved in an accident.
Crash contact point optimisation combines passive and active safety. Optimising the car’s orientation in a collision, i.e. where it will be hit, makes it possible to reduce the impact of the crash, and optimise the benefits of airbags and seatbelts and the efficacy of structural means of absorbing crash energy.
AstaZero’s location, with warm summers and cold winters, enables year-round testing. One of the benefits is that the reliability of the active safety systems can be tested under different climate conditions depending on requirements and type of test. One area where climate conditions can be of major importance is in the use of batteries in electric cars, and how they are affected by very high or low outdoor temperatures. Furthermore, different weather conditions, such as direct sunlight, wet roads or reflective snow, influence the performance of the sensors and thus in turn also the reliability of active safety systems.