The eSafety Initiative and the Integrated Safety Program
Every year, about 45 000 people die and 1.5 millions people are injured in traffic accidents in Europe. These figures imply that one person out of every 200 European citizens is injured in a traffic accident each year and that around one out 80 European citizens die 40 years short of their expected lifetime.
It is known that the great majority of road accidents (about 90-95%) are caused by human error (Treat, et al., 1979). More recent data has identified inattention (including distraction, "looked but did not see" and falling asleep at the wheel) as the primary cause of accidents, accounting for at least 25% of the crashes (Wang et al., 1996).
Road safety is thus a major European health problem. In the "White Paper on European Transport Policy for 2010", the European Commission declares the ambitious objective to reduce by 50% the number of fatal accidents on European roads by 2010 (European Commission, 2001). While conventional vehicle safety measures (e.g. seatbelts and airbags) have contributed significantly to the reduction of accidents in the last decades, their contribution is reaching its limits and currently further improvement is difficult to achieve at a reasonable cost.
Today, the development of new Advanced Driver Assistance Systems (e.g. collision avoidance-, lanekeeping aid- and vision enhancement systems) offers great potential for further improving road safety. In order to accelerate the research and development and deployment of these technologies, the eSafety initiative has been set up. This is a European joint public-industry initiative aimed at promoting the development and deployment of Intelligent Road Safety Systems.
The AIDE Integrated Project (IP) has been set up to address HMI issues within a general European joint effort towards the large-scale deployment of Intelligent Road Safety Systems and, ultimately, a significant reduction of road accidents.
HMI design for maximising the safety benefits of new Advanced Driver Assistance Systems (ADAS)
Today, a wide range of Advanced Driver Assistance Systems (ADAS) are being developed for enhancing the driver's perception of the hazards, and/or partly automating the driving task. These include speed alert, lane support/blind spot detection, automated safe following, pedestrian detection, vision enhancement and driver impairment monitoring. These systems have great potential for reducing accidents, in particular the great portion related to human error (European Commission, 2002).
The safety impact of these systems depends will to a great extent be determined by their interaction with the driver. For example, in order to efficiently support the driver in avoiding crashing into a front obstacle, it is crucial that the warning/feedback given by the system intuitively generates the appropriate response (e.g. an avoidance manoeuvre).
New technologies, exploiting new concepts for driver-vehicle interaction in multiple sensory modalities (e.g. visual, tactile and auditory), offer great potential for maximising the potential safety benefits of ADAS. Research and development on how to best exploit these possibilities to maximise the efficiency of ADAS is urgently needed.
Moreover, it is well known that the introduction of new safety functions may induce longer-term changes in driver behaviour.
This type of behavioural change, often referred to as behavioural adaptation, may significantly affect the actual (as compared to the expected) safety benefits of a safety measure, both in positive and negative directions (OECD, 1990). Behavioural effects demonstrated for ADAS include system over-reliance on in-vehicle safety technologies resulting diversion of attention from the driving task and safety margin compensation.
However, the mechanisms underlying these effects are largely unknown and a model for predicting them does not exist. Finally, the potential safety impact of an ADAS ultimately depends on its market penetration rate and whether it is actually used by drivers.
Here, the human-machine interface is of crucial importance; annoying system behaviour (e.g. nuisance warnings) will lead to drivers simply abandoning the system, which hence obviously looses its potential safety benefit.
HMI design for minimising workload and distraction imposed by In-vehicle Information Systems (IVIS)
In addition to Advanced Driver Assistance Systems, a growing number of In-vehicle Information Systems (IVIS) are being introduced in modern vehicles.
By contrast to ADAS, these systems provide services not directly relevant for the primary driving task and thus impose a secondary tasks on the driver. Moreover, the in-vehicle use of portable computing devices e.g. hand-held mobile phones and portable digital assistants (PDAs), often referred to as Nomad devices, is increasing rapidly.
These systems have great potential for increasing mobility and comfort. For example fleet management systems enhance the efficiency of work in the freight industry and road-and traffic information systems potentially facilitate the quality of life for the commuter. However, information systems in vehicles may also compete with the primary driving task for the driver's attention and hence induce dangerous levels of distraction and workload. The safety risks of IVIS are well known, in particular the case of mobile phones.
Given this critical safety impact of mobile phones alone, the introduction of additional information functions such as email, internet access, navigation aids, road and traffic information raises obvious safety concerns. Nomad devices are not even designed for use while driving and thus major potential future in-vehicle distractors. The design of the human-machine interface of IVIS and nomad devices is of key importance for minimising the workload and distraction that they impose on the driver. Methods and criteria are needed to validate these systems with respect to their potential negative safety effects.
Towards a unified HMI solution for integrating ADAS and IVIS
In addition to the safety issues associated with the individual systems described above, the proliferation of complex in-vehicle functions itself poses a further challenge for the design of the driver-vehicle interface. Figure X below illustrates the situation facing vehicle system HMI designers in the near future.
Examples of various ADAS and IVIS interacting with the driver in a future vehicle.
Click to enlarge!
It is clear from Figure above that the various systems interacting with the driver cannot be implemented independently. The most obvious reason for this is that such a large number of separate HMI devices would simply not fit into the vehicle cockpit.
Moreover, conflicting information from different systems may distract, overload, confuse and annoy the driver, thus causing problems that did not exist for the systems in isolation.
Moreover, behavioural changes in response to a combination of systems may be very different from responses to the systems in isolation. Thus there is a strong need for a unified human machine that integrates the different systems into functioning whole, resolving conflicts between different functions and taking into account their aggregate effects.
Some key features of such an integrated HMI would include:
- Multimodal HMI devices shared by different systems (e.g. head-up displays, speech input/output, seats vibrators, haptic input devices, directional sound output)
- Centralised intelligence for resolving conflicts between systems (e.g. by means of information prioritisation and scheduling).
- Seamless integration of nomad devices into the on-board driver-vehicle interface.
- Adaptivity of the integrated HMI to the current driver state/driving context