- Email: [email protected]
Presenting information to the driver
Presenting information to the driver M. FOWKES The preseatAtlon of information to the driver is assessed f~om a historical perspective, highlighting factors that have influenced current practice, and therefore expectation of insmunent panel displays. The driveCs use of these displays is then analysed from a systems viewpoint, idemifying important factors which influence the visibility and legibility of displays. The development at the Motor Industry Research Association of a vehicle simulator is reported, and the manner of its use to study the important factors above is d ~ An examination of current practice, legislation and human factors research suggest that future implementation of on-board vehicle electronic displays could assist drivers by simplifying the driving tasks, ie reducing current levels of visual inspection tasks and information processing.
Keywords: dashboards; irmmunentation; ergonomics; legibility; display devices (computers). Information on the status of various vehicle systems has been presented to drivers in the form of visual displays since the advent of the automobile. Clearly, on the earliest vehicles the information which could be displayed was limited by the transducers which were available, and information was only provided that was thought to be important for the driver to be aware of when driving. Vehicle road speed, fuel level and engine (coolant) temperatures in particular were amongst the first parameters to be commonly displayed. This information could then be interpreted by the driver to suggest when a dangerous or undesirable condition had been reached so that appropriate action could be taken. In other words vehicle visual displays were provided to present information on the status of systems that the driver could not perceive easily or accurately with his own unaided senses. This general statement could as easily apply to modern vehicles currently in use. However, recent developments in solid state electronics have made possible the production of visual displays that can present information in a novel form under automotive environmental conditions. The wider application of these devices in the future therefore heralds potentially large changes in the presentation of information to the driver. While these displays offer potential benefits to the manufacturer and the driver, their suitability for use under these vehicle environmental conditions by the driving population must yet be established. This paper seeks to examine this suitability after briefly reviewing the historical development of vehicle visual displays. The overall system of driver-vehicle-environment will then be examined as well as specific problem areas and the means by which they can be studied. Finally the possibilities for future developments in display applications are discussed.
cated. However until recently the basic means of presenting this information visually have changed little. Early automobiles mostly used circular moving pointer and fixed scale mechanical gauges as these were readily manufactured given the engineering abilities available at that time. The displays were normally mounted in the front bulkhead of the vehicle, conveniently near to both the engine and transmission and within the forward view of the driver. As vehicles became more refined the amount of information presented gradually increased. Figure 1 shows the instrument panel of a 1927 Bentley which may be thought to represent the state-of-the-art up to 1930. Many separate instruments, each specific to a vehicle function, are spread across the full width of the vehicle, each seemingly having a different design of pointer, scale marking, type of motion etc. Over the next 20 years of vehicle development little changed in the technology of presenting visual information, and moving pointer and fixed scale electromechanical instruments were universally used. However the manner in which these instruments were mounted in the vehicle did change. Gradually, the instrument
HISTORICAL DEVELOPMENT OF V E H I C L E DISPLAYS Over the last 80 years vehicle systems and measuring transducers have in general become more sophisti"'
The author is at the Motor Industry Research Association, Watling Street, Nuneaton, Warks CVIO OTU, UK.
DISPLAYS. OCTOBER 1 9 8 4
+.,
Fig. 1 Instrument panel of a 1927 Bentley
0141-9382/84/050209--09 $03.00 ~ 1984 Butterworth & Co (Publishers) Ltd
215
panel became more compact and major displays were presented directly ahead of the driver so that they were normally viewed through the steering wheel. Only supplementary gauges were mounted outside this area. Also the use of tell-tale lights increased to present simple information either as system status or warning functions. As this latter development grew, the need to identify each separate light by colour, symbols or lettering became apparent in order to prevent confusion. In short, this more integrated instrument panel helped to reduce the amount of time the driver's view was distracted from the road. From the manufacturer's point of view this form of instrument panel also gave more flexibility in interior design as well as production economies, due to more ready use of sub-assemblies.
controls, indicators and tell-tales, as for example through the work of the International Standards Organisation ~. Its standard for road vehicle instrumentation used the symbol of Fig. 4c for the choke as the one most commonly understood and therefore most desirable as a basis for standardization. However, it is a long process for the 'standard' to be universally accepted, by first of all appearing on new vehicles and then gradually taking over the market, as vehicles are replaced.
These gradual changes in the detailed format of displays were undoubtedly influenced by the large amount of research work performed on visual display design throughout the post-war period. This interest arose particularly for aeronautical applications, and examined many aspects of the effectiveness of reading displays that had obvious parallels in vehicles use. A few examples of the aspects studied were different combinations of scale and pointer, direction of movement of dials and pointers, fonts of characters and use of symbols. This research interest in displays design was paralleled in the production of unusual vehicle displays by some manufacturers. Horizontal rotating bar and fixed pointer and moving scale speedometers were both used in different vehicles, for example. However, such innovations remained much in the minority and circular form electromechanical instruments predominated. Figures 2 and 3 show the instrument panels fitted to two typical production cars of the late 1960s, a Fiat 128 and a Honda $800 respectively. These two examples show more simplified and clearer presentation of information on much the same vehicle functions as on the instrument panel of the Bentley approximately 40 years earlier. Vehicle legislation had by this time also begun to exert a minor influence on display design as national standards were established for the minimum information which needed to be displayed in vehicles sold in a particular country. However, little international agreement between these requirements was apparent. This was particularly true in the field of symbols or lettering used to identify vehicle functions on displays, controls or tell-tales. Figure 4 shows three different symbols used for the choke control or tell-tale which were used on vehicles sold in the same international markets. Surveys of drivers at that time showed that what was understood from these symbols differed from country to country. Whilst a symbol was almost universally recognized in its country of origin, it was found to be unknown in others. It should therefore be apparent that population expectations of what a display means are based on prior experience and can be powerful influences on the success of a visual display. Clearly some form of standardization was required. The major developments in vehicle visual displays from 1970 to the present day lie in the area of display technology. International legislation has begun to harmonize the presentation of symbolic information for 216
Fig. 2 Instrument panel o f a 1968 Fiat 128
Fig. 3 Instrument panel o f a 1967 Honda $800
a
b
I ,1
e
Fig. 4 Three symbols o f choke control and displays; a - - UK; b - - France; c - - F R G (ISO 2575)
DISPLAYS. OCTOBER 1984
In recent years the development of many forms of electronic display technology has made a range of components available for use in vehicles. Initially these were small discrete displays dedicated to supplying supplementary information as an additional component to be mounted alongside conventional electromechanical instrument packages. Commonly these devices gave a digital display of a variable, the most common application being a clock. Further developments in display technology have now reached a point where complete instrument panels are produced in solid state electronics. Figure 5 shows a current production vehicle, an MG Maestro, fitted with such a panel. This current application of electronic displays therefore uses new technology to replace earlier electromechanically displayed functions on an item-for-item basis. Specific areas of these electronic instrument panels are uniquely used for displaying a specific function. However, the choice of format for this information display is made far wider than with conventional instruments. What then for the future? Currently, the implementation of electronic displays has affected only a small proportion of all new vehicles. Solid state electronic instrument panels have been fitted to vehicles at the higher price end of the market, or as top of the range options to more common models. However it is virtually certain that within a short time these displays will be fitted to the majority of new vehicles. After all the attractions to the designer and manufacturer are understandable. Electronic displays will integrate well with the increasing importance and frequency of the electronic management of a number of vehicle systems. Such system control functions will also provide the sensing and processing required for information display and could also provide the processing back-up for even more flexible presentation of that information. These electronic systems as a whole therefore offer advantages over conventional electromechanical systems in efficiency, potentially reduced weight and size, as well as increased flexibility. Also as research and development continue into the components themselves, increased reliability and production efficiencies will help to highlight the economic advantages of solid state electronics. In short, the electronic display will soon become the norm rather than the exception.
ERGONOMIC DESIGN What of the driver? He or she will clearly be affected by this potentially radical change in the means by which he is kept informed of vehicle system performance. It is obviously the role of the human factors specialist to assess the implications of this imminent change in vehicle design and to provide information on how this interface between driver and display, may be most successfully designed. This ergonomic assessment can be said to be at two levels, the general and the specific. First, the presentation of information to the driver should be re-evaluated in the light of new display technology. The driver's use of display should be assessed in the context of current vehicle and environmental conditions, and those likely in the future, to ensure that all aspects of the complex interaction between man and the machine are examined. Second, individual factors which are important to the total system need to be identified and classified. Specific problem areas can then be assessed under controlled conditions. In this manner the relative importance of these potential problems can be evaluated, and suitable solutions determined. A n a l y s i s o f the s y s t e m The visual display has been described as an interface between man and machine, driver and vehicle. This terminology is used to describe a site or location where information is passed between two components of a system. In present discussion these are two distinct aspects that affect the efficiency of this information transfer. These may be termed visibility and legibility. A display that is visible may be seen correctly by the viewer. A display that is legible however is also correctly understood by the viewer. As an example, an English book read by a non-English reader may be visible to him but will not be legible. However, a display that is not visible is hardly likely to be legible! These two aspects of the efficiency of the driver-display interface form the basis of the system analysis and therefore are critical to display design. There are four groups of factors which interact with one another to affect this success: they can be categorized as human, environmental, display and task factors. These are portrayed diagrammatically in Fig. 6. This analysis of the important factors contributing to the effectiveness of the visual display interface should not be though to be exhaustive, but should be viewed as a simplified model outlining the most crucial factors. A brief discussion of these individual factors is given in Tables I to 4. Specific ergonomic studies Following the analysis of the system which is described above, specific areas of factor interaction may be identified which are though likely adversely to affect display visibility or legibility. These potential problem areas are best examined in detail under controlled conditions.
Fig. 5 Instrument panel o f a 1983 MG Maestro
DISPLAYS. OCTOBER 1984
Recent research work performed at MIRA was undertaken to study several specific areas of interaction of these factors under laboratory conditions. This study was made possible by the construction of a vehicle 217
/
• • • •
Environment Illuminationleveland spectrum Vibration Spaciallocationof driver and display Glare SOurces
Human factors s Aoe/se~ = Visual acuity q i Colourvision • Fatigue • Expectancy • Night-timevisior • Language • Reaction speed • • • • •
[ Displayfactors [ I• Characterangularsize I J• Characterluminance/ J contrast I J• Character generationJ la Font/graphic design I J • Colour J
Environmental
factors
I Visual display J Visibility/legibility I
Task / Levelof workload Forward or reverse vehicle motion Display obscuration Type of information required Vehicle speed
Fig. 6 Driver.display interface Nctors simulator, designed to be able to vary these levels of factors that were identified as important for study. In particular, the following factors are controlled within the simulator.
•
Ambient illumination measured at the instrument panel, variable between 10--80 000 Ix; Correct spectral balance for chosen illumination level; Range of driver/display spatial locations to represent all road vehicle values; Facility to present glare sources.
Human factors • Adjustment to suit the range of sizes in the drivers population; Facility to measure the performance of the subjects in reading displays; Display factors • Ability to house a wide range of size and types of displays, with a variety of display characteristics; Facility to measure the photometric performance of each display; Task factors
• Facility to incorporate secondary tasks within the simulator.
The simulator was constructed to these parameters and ~ncorporated a total 1.35 kW of artificial illumination
Table 1. Human factors
Human Factor
Description
Age
The driving population ranges in age from 17 to more than 90 years old. Many aspects of human ability suffer impairment with increased age. Visual performance also follows this trend. Visual acuity and colour sensitivity are two examples, and their age effects are discussed below. Accommodation, or the ability of the eyes to focus at different distances, is affected by ageing. The nearest focusing distance for the normal 16-year old can be 80 ram. This may increase to 250 mm at 45 and to I metre at 60. RESULT: In specifying character size for displays used by the full range of drivers in the population, particular note must be taken of those over 60 years both in terms of acuity and accommodation.
Sex
The main characteristic of importance which is related to sex is colour vision occur mainly in the male population. RESULT: See colour vision below.
Visual acuity
Visual acuity can be described as the ability of the eye to resolve fine detail. The effect on performance depends on the size of target and its brightness. Consequently the level of illumination present and the contrast between target and background are important contributory factors in determining visual performance. This ability is age-dependent, ie with increased age visual acuity decreases. Some sources suggest performance decrements of 50 per cent for the population of 40+ compared to the 20-40 age group. Therefore drivers over 40 may require up to twice the character size as those younger at threshold values. Visual acuity is also impaired under night time conditions. RESULT: Again the age effects of acuity require a close consideration of the needs of the older driver.
Colour vision
Deficiencies in the colour discrimination ability of the population occur particularly in the male population. There are many forms of colour vision deficiency but the most common are difficulties in discriminating between red and green or blue and yellow. In addition colour discrimination generally deteriorates with increased age. RESULT: Once more the requirements of a portion of the population dictate the requirements for the overall population. Colour combinations to avoid are red/green and blue/yellow due to confusion problems.
218
DISPLAYS. OCTOBER 1984
mounted on multi-way adjustable frames within the simulator to give the desired levels and type of illumination. The simulator itself was constructed as a framework which represented a vehicle compartment with front bulkhead, instrument panel, driver's door and roof. In addition a steering wheel and column could be added. The driver's seat was adjustable in horizontal, lateral and vertical directions relative to the instrument panel which was also adjustable for angle of rake and lateral position. The simulator was housed in a room blacked out to external daylight. Three specific areas were studied using this simulator - - a m b i e n t illumination and glare; character size and
angle of view; and analogue or digital presentation of information; In all of these studies, small groups of subjects selected from the driving population were trained in the experimental tasks before their performance in using the display was measured. These performace measures did vary between studies but speed and accuracy of reading the display were the most important experimental response measures. These were chosen as they had been shown to provide a sensitive measure of the factors under study. Speed of reading was measured
Fig. 7 Vehicle simulator test rig by controlling the exposure of test displays to the subjects by means of a mechanical shutter. The simulator in this form is shown in Fig. 7. The simulator also has the facility to provide a secondary tracking task which is
Human Factor
Description
Fatigue
Fatigue in the context of driving is an important factor to consider where the question of crucial event information is concerned. Fatigue effects upon the driver are shown, in increasing order of importance; in inattention to individual components of the driving task, erratic visual search behaviour, and finally short periods of unconsciousness or 'microsleep'. A display cannot eradicate these effects, of course, but crucial event information should attract attention in an attempt to counter the less severe fatigue effects. RESULT: All crucial event information should be 'attention' attracting to the driver.
Expectancy
The driver population as a whole has expectancies of the information and format of information given by vehicle displays. In terms of general format the common fixed scale-moving pointer display has been present in practically all vehicles for major display functions. Hence the driving population expect this format, and are trained to use it. Colour coding such as 'red' warning lights, 'blue' main headlamp beam etc are accepted and standardized. However symbology for functions differs between populations. RESULT: Novel display formats should be evaluated for their potential acceptability by target populations before use. Careful note of population expectations should be undertaken before design for any particular market.
Night-time vision
The eye uses two mechanisms of the retina to detect light. These are described by their functional units in the retina, the rods and cones. The cones are primarily sensitive to the wavelength of light and therefore colour, and are predominant around the optical axis of the eye. The rods are more sensitive to the amount of light present and are mainly situated at the periphery of vision. RESULT: Colour discrimination at the periphery of vision during night-time conditions is seriously impaired. Therefore crucial event information is required to be sufficiently attention seeking to attract the central gaze of the eye which is more sensitive to colour.
Language
This applies mainly to alphanumeric displays which could present information in a written form. Intelligibility should be assessed for all countries where the vehicle will be used. RESULT: Consider target markets in providing coded information.
Reaction speed
There is variation amongst the driving population in almost every respect, including response speed. In terms of visual display this requires consideration of the format and content of information so that it is easily assimilated by the slowest responders. RESULT: Evaluation of novel designs requires testing by a wide range of subjects from the population.
DISPLAYS. OCTOBER 1984
219
Table 2. Environmental factors
Environmental factors Illumination level and spectrum
Description The illumination level found at the instrument and spectrum panel surface of a passenger vehicle is of course variable with the time of day and season of the year. For European driving it may be estimated that the range of levels may be from 0.01 Ix (moonlight) to 60 000 Ix (direct side window solar illumination). The spectral qualities of the illumination present in a vehicle can be variable. Daylight conditions, where the sun is the light source, provide wide-band illumination even though this can take on pronounced casts depending upon solar altitude. However, more veiling illumination is provided by narrow-band emitting artificial road lights. RESULTS: High levels of direct illumination can cause wash-out of active displays, and colour of the incident light can greatly influence visibility, particularly of narrow-band emitting warning lights. Evaluation techniques for displays must accurately recreate both level and spectra of representation vehicle ambient illumination ranges, taking careful note of any specific operating conditions of special purpose vehicles.
Vibr~ion
Vibration induced in both the driver and any visual display in the instrument panel is created by complex interaction of the road surface, tyre and suspension characteristics. The result is a variable acceleration/freqency pattern during driving. Typical values noted for passenger care are 0.16 ms -z at 4 Hz and 0.1--0.3 ms -2 at 11 Hz. If a vehicle is due to operate under environments where more extreme vibration conditions may occur then of course more substantial effects may be seen. RESULTS: If vibration levels are likely to be severe then careful consideration of this factor is required perhaps to increase subtended angle of characters.
Spatial location of driver and display
The fascia/seating package of a vehicle imposes driver and display basic limitations on the display factor of subtended angle. For most normal passenger car packages a typical display viewing distance is 800 ram, and this has been used in this study. The angle of view of the display is also controlled by the manner in which the display is mounted in the fascia relative to the direct line of sight to the driver. RESULT: Current packaging suggests 800 mm as an acceptable viewing distance for the character sizes use in current visual displays, ideally placed so that the deviation required from the normal line of sight is minimized and that the displays are viewed normally.
Glare sources
These may be either natural sources (ie sun, either viewed directly or reflected) or artificial sources (vehicle headlights or roadside lights). Evidence from experiments suggests that current practice glare levels and display characteristics a not sufficiently near threshold values for this factor to be a problem. RESULT: If the level of glare is exceptional due to predicted operational conditions then further consideration should be given to display visibility factors.
analogous to that found while driving. This consisted of a pointer controlled by a steering wheel tracking a path drawn on a scroll pulled mechanically past a viewing aperture at windscreen level. Subjects were asked to read values displayed on their electronic instruments whilst performing the tracking task. A performance measure was derived in this case from the accuracy of both display reading and tracking ability. This arrangement is illustrated in Fig. 8. Computer-generated and controlled secondary tracking tasks are under development which can replace the mechanical arrangement used previously.
FUTURE DEVELOPMENTS Research data is available on many of the areas identified in the analysis given above. Indeed design guidelines have been developed from this information by a number of experts in this field. However, this should not suggest that all questions are answered and all problems solved. In an area of rapidly developing 220
technology some reassessment of current guidelines in the light of future innovations is required, quite apart from filling in those gaps that exist in our current knowledge. For these reasons laboratory-based research that uses facilities such as the MIRA vehicle simulator will continue to be required. This could then lead to a restatement of desirable design guidelines in the form of a specification based upon legibility test and performance criteria. This latter approach has been suggested by Gorrell e for assessing the legibility of video monitors and used a refined set of procedures for human factors testing. There is therefore a possibility that a similar approach could be developed for vehicle displays, which would undoubtedly be a more difficult undertaking due to the more complex nature of display use in a vehicle. However, there are also wider questions to be posed for the future presentation of information to the driver using electronic displays. If electronic displays are to be fitted to the majority of vehicles in the near future then it is pertinent to assess
DISPLAYS. OCTOBER 1984
Table 3. Task-related factors
Task-related factor
Description
Level of workload
The level of workload current on a driver during a journey is very variable. Quiet, relatively 'straight' rural roads offer relatively little workload in comparison to a congested town centre. In the context of visual displays this may seem that at times during a journey a driver has little or no spare capacity to view his instrument panel mounted display. RESULTS: The 'worst case' road situation which provides a high workload for the driver can completely interrupt the subsidiary task of periodically viewing the displays. If crucial event information (such as oil pressure failure) is to be conveyed to the driver successfully under these situations then it has to be of a form that can attract his attention, possibly by using a nonvisual mode.
Forward or reverse
This is the most obvious task-related factor to provide an interruption to the driver's ability to inspect his visual displays. When reversing a vehicle, the driver's view is normally directed rearwards and therefore away from the instrument panel. RESULT: As with 'level of workload' display reading tasks are potentially interrupted. Similar comments therefore apply. If crucial event information is to be given it must attract the driver's attention.
Type of information
A crucial factor is the appropriate determination required of type and format of information display. Various classifications have been advanced to describe the type of information required by a user from a display. Briefly, these can be described as quantitative, qualitative, check, and status/warning information. These are the principal visual display uses in a vehicle. Although with the future of electronic displays in mind, a fifth category which is the more detailed descriptive information provided by alphanumeric/symbolic displays may also be included. RESULT: In any assessment of a complete instrument panel design careful consideration should be given to each individual displayed item of information. What form of information is required from the display and how is this information coded to the driver in relation to all other displayed functions?
Vehicle speed
A special example of increased workload with higher vehicle speeds. Under motorway conditions, for example, the driver's visual search of the road is increasingly further ahead of the vehicle. Any distraction from this to inspect the instrument panel requires not only a substantial re-accommodation of the eye, but also causes a greater penalty in terms of road distance covered without the driver's view being directed to the road. RESULT: At high vehicle speeds the need for speed of reading from visual display is very much greater.
the radical departures from current practice that will be available to vehicle manufacturers and indicate which offers most promise for the driver. The current production status of electronic instrument panels has been described in a replication of earlier electromechanical implementations on an item-for-
Array of 3 fluorescent tubes
miler
'
~ ~/.~
/
/
Hopper collecting road pattern
Fig. 8 Vehicle simulator test rig
DISPLAYS. OCTOBER 1984
~ " ~ M a t t black roof section aperture ~ f o r tracking task ta! / Subject's ocular ~_~ . . . ~ . . ~ position
item basis. Therefore, while the medium and form for presenting information has changed, the basic task for the driver remains the same. He must still inspect the various displays on the instrument panel, process the information when received and then formulate and carry out an appropriate response. Specific human factors appraisals, sensibly constructed design guidelines and legibility tests may ensure that a full electronic instrument panel is at least as visible and legible as its earlier electromechanical counterparts. This situation does of course have the advantage that the driver's expectations of the form and functions of the instrument panel are not compromised. In short, to the driver the appearance of the instrument panel are not compromised. In short, to the driver the appearance of te instrument panel has changed little. However, this does mean that the visual inspection and information processing required by the driver remains largely the same. The increase in complexity of information presented on the instrument panel of vehicles has also been described. Part of the reason for this increase may be credited to the instrument panel's importance in
221
Table 4. Display factors Display factors
Description
Character angular size
The minimum acceptable size of a display character (normally expressed in minutes of arc) size is most closely related to the visual acuity of the driver and the location of the display. Taking note of character contrast, ambient illumination level, character font, human factors etc these have been accounted for. The minimum value suggested for accurate reading is: RESULT: 20 min of arc (ie 4 mm viewed at 800 mm)
Character luminance/contrast
In general terms, the legibility of any display is increased with greater contrast between symbol and background. In vehicle use this factor influences legibility as much as any other. Given many sources of research, and taking the minimum character angular size noted above, it is suggested that the value below should give acceptance legibility to the driving population. RESULT: At least 8:1; however, colour contrast is still in need of further research.
Character generation
For alphanumeric information in particular, two forms of character generation are available, segmented or dot matrix designs. Various studies have been performed in a variety of technologies attempting to assess the relative legibility of these formats. RESULT: The more segments present in a segmented display, ie seven segments and above then the greater the flexibility of character which may be presented. However, in most comparisons if contrast ratios a r e kept to acceptable levels (as given above) little difference is found between seven segment and various dot matrix characters (see below).
Font/graphic design
The graphic design or font of alphanumeric characters has been investigated at length by many researchers, much of the work referring to printed characters on conventional displays. This has evaluated many aspects of stroke-width and width-height ratios. RESULT: Many studies have shown remarkably small differences between fonts evaluated in dot matrix form: ASCII, Lincoln/MITRE and NAMEL are amongst those shown to be very similar in legibility tests. However, it is suggested that to display a range of non 'alphanumeric' symbols greater flexibility of character generation is needed than that available on a sevensegment character.
Colour
Colour vision deficiencies within the male population require that the use of colour within display designs in general should be controlled. There are, of course, existing legal requirements for colours used for displaying certain functions. Separating status and warning lights may be an important consideration, to provide position as well as colour coding of types of information along with any associated symbol, for particular vehicle functions. RESULT: It is important to consider display colour problems with particular reference to colour discrimination, combination of colours and symboloy.
Others
Current advances in electronics enable us to consider which sensory modalities should be used in specifying displays. Previous comments have suggested that crucial event displays (such as 'low oil pressure', 'overheating' and 'no fuel' warning require immediate action on the part of the driver. For a visual display to have the required attention attracting characteristics then briefly flashing 'emergency' lights or auditory signals could be used as a back-up to 'request' attention to the status/warning displays. RESULT: Modern electronics will enable a range of auditory messages to be presented to the driver either in symbolic (ie buzzer/bell etc) or verbal form to require attention to a particular vehicle function. The integration of audio and visual presentation of information could prove a significant improvement to the notification of vehicle system state by simplifying the visual load of the driver.
the interior trim of a vehicle, and therefore its use in establishing different models within a vehicle range. Generally speaking, the more expensive the model within a range, the greater the complexity of instrument panel information. However, the need for high current levels of information on display is questionable. Today's congested road conditions require the full attention of the driver to be directed towards the events in the external road environment in order that the journey may be completed safely. This may mean either 222
that the driver regularly inspects all displays provided and is interrupted in the driving task, or, at the other extreme, that he ignores the displayed information and concentrates solely on the immediate road and traffic conditions. Clearly the actual behaviour of a driver between those two extremes is controlled by other factors, principally the information processing load caused by road and traffic conditions. When traffic is extremely busy the driver will largely ignore displayed information and vice versa. A second consideration
DISPLAYS. OCTOBER 1984
which further confuses this discussion is that of the interpretation that the driver makes of this displayed information. Probably all drivers would grasp the significance of a fuel level instrument showing 'E', 'Empty' or '0'. However, how many drivers would have the technical understanding of vehicle functions to grasp the significance and possible consequences of a sudden drop in oil pressure? Secondly, how many drivers would then select an appropriate course of action to avoid the undesirable outcome. Perhaps it is more important then to use the flexibility of vehicle electronic systems and displays to lighten the information processing load of the driver, and to assist the technically inexperienced members of the population understand the information displayed. The use of dot-matrix forms of a number of display technologies has made possible a redefinable display panel which can produce both alphanumeric and graphical forms of information as desired. Prototype installations of this form of display have been shown in recent years fitted to experimental vehicles. The most advanced of these have used a single display panel that gives a simple display of a minimum number of functions, for example speed, fuel level and elapsed mileage only. However, the microprocessor controlling the display can be instructed by the driver to display more detailed information on specific functions when required, eg oil pressure, oil temperatures etc. This new information displayed on the single central display panel which after a short period returns to the original basic information. In addition these interactive processors perform individual system status checks on all monitored functions. Dependent upon the amount of processing ability these checks can either notify the driver of 'warning' conditions, ie safe limits exceeded, or 'caution' conditions, ie safe limits approaching. Clearly these systems could be developed to even give simple advice as messages when warning or caution conditions are reached. These would be obviously tailored to alleviate the problems of the non-technical drivers. It is now also feasible to present some of these alphanumeric messages as spoken messages. While there are obvious detailed problems of auditory legibility to be overcome, this dis-
DISPLAYS. OCTOBER 1984
Fig. 9 Cathode ray tube display play mode should not be neglected, and its integration with the visual displays used should be considered in detail. The future could therefore see a simplified presentation of constantly displayed information which on demand from the driver or internal diagnostic functions could present a wealth of detailed information if required. Simple messages would be able to advise the driver of critical system conditions or failures in such a way that they were universally understood. This however is some way in the future. On-board diagnostics and message centre form instrument panel displays are in their infancy for mass-produced vehicles. Figure 9 shows a CRT-based display illustrating how this approach may appear to the driver. It remains to be seen whether this is the likely appearance of future vehicle displays, and whether human factors guidelines are used to develop systems which are a help to the driver and not a hindrance.
References 1 ISO 2575 'Road v e h i c l e s - symbols for controls, indicators and tell-tales' 2 GorreH, E.L. 'A human engineering specification f o r legibility of alnumeric symbology on video monitor displays'DCIEM Tech Rep 80-R-26 (Defence and Civil Institute of Environmental Medicine, Depart ment of National Defence, Canada, 1980)
223
Keywords: dashboards; irmmunentation; ergonomics; legibility; display devices (computers). Information on the status of various vehicle systems has been presented to drivers in the form of visual displays since the advent of the automobile. Clearly, on the earliest vehicles the information which could be displayed was limited by the transducers which were available, and information was only provided that was thought to be important for the driver to be aware of when driving. Vehicle road speed, fuel level and engine (coolant) temperatures in particular were amongst the first parameters to be commonly displayed. This information could then be interpreted by the driver to suggest when a dangerous or undesirable condition had been reached so that appropriate action could be taken. In other words vehicle visual displays were provided to present information on the status of systems that the driver could not perceive easily or accurately with his own unaided senses. This general statement could as easily apply to modern vehicles currently in use. However, recent developments in solid state electronics have made possible the production of visual displays that can present information in a novel form under automotive environmental conditions. The wider application of these devices in the future therefore heralds potentially large changes in the presentation of information to the driver. While these displays offer potential benefits to the manufacturer and the driver, their suitability for use under these vehicle environmental conditions by the driving population must yet be established. This paper seeks to examine this suitability after briefly reviewing the historical development of vehicle visual displays. The overall system of driver-vehicle-environment will then be examined as well as specific problem areas and the means by which they can be studied. Finally the possibilities for future developments in display applications are discussed.
cated. However until recently the basic means of presenting this information visually have changed little. Early automobiles mostly used circular moving pointer and fixed scale mechanical gauges as these were readily manufactured given the engineering abilities available at that time. The displays were normally mounted in the front bulkhead of the vehicle, conveniently near to both the engine and transmission and within the forward view of the driver. As vehicles became more refined the amount of information presented gradually increased. Figure 1 shows the instrument panel of a 1927 Bentley which may be thought to represent the state-of-the-art up to 1930. Many separate instruments, each specific to a vehicle function, are spread across the full width of the vehicle, each seemingly having a different design of pointer, scale marking, type of motion etc. Over the next 20 years of vehicle development little changed in the technology of presenting visual information, and moving pointer and fixed scale electromechanical instruments were universally used. However the manner in which these instruments were mounted in the vehicle did change. Gradually, the instrument
HISTORICAL DEVELOPMENT OF V E H I C L E DISPLAYS Over the last 80 years vehicle systems and measuring transducers have in general become more sophisti"'
The author is at the Motor Industry Research Association, Watling Street, Nuneaton, Warks CVIO OTU, UK.
DISPLAYS. OCTOBER 1 9 8 4
+.,
Fig. 1 Instrument panel of a 1927 Bentley
0141-9382/84/050209--09 $03.00 ~ 1984 Butterworth & Co (Publishers) Ltd
215
panel became more compact and major displays were presented directly ahead of the driver so that they were normally viewed through the steering wheel. Only supplementary gauges were mounted outside this area. Also the use of tell-tale lights increased to present simple information either as system status or warning functions. As this latter development grew, the need to identify each separate light by colour, symbols or lettering became apparent in order to prevent confusion. In short, this more integrated instrument panel helped to reduce the amount of time the driver's view was distracted from the road. From the manufacturer's point of view this form of instrument panel also gave more flexibility in interior design as well as production economies, due to more ready use of sub-assemblies.
controls, indicators and tell-tales, as for example through the work of the International Standards Organisation ~. Its standard for road vehicle instrumentation used the symbol of Fig. 4c for the choke as the one most commonly understood and therefore most desirable as a basis for standardization. However, it is a long process for the 'standard' to be universally accepted, by first of all appearing on new vehicles and then gradually taking over the market, as vehicles are replaced.
These gradual changes in the detailed format of displays were undoubtedly influenced by the large amount of research work performed on visual display design throughout the post-war period. This interest arose particularly for aeronautical applications, and examined many aspects of the effectiveness of reading displays that had obvious parallels in vehicles use. A few examples of the aspects studied were different combinations of scale and pointer, direction of movement of dials and pointers, fonts of characters and use of symbols. This research interest in displays design was paralleled in the production of unusual vehicle displays by some manufacturers. Horizontal rotating bar and fixed pointer and moving scale speedometers were both used in different vehicles, for example. However, such innovations remained much in the minority and circular form electromechanical instruments predominated. Figures 2 and 3 show the instrument panels fitted to two typical production cars of the late 1960s, a Fiat 128 and a Honda $800 respectively. These two examples show more simplified and clearer presentation of information on much the same vehicle functions as on the instrument panel of the Bentley approximately 40 years earlier. Vehicle legislation had by this time also begun to exert a minor influence on display design as national standards were established for the minimum information which needed to be displayed in vehicles sold in a particular country. However, little international agreement between these requirements was apparent. This was particularly true in the field of symbols or lettering used to identify vehicle functions on displays, controls or tell-tales. Figure 4 shows three different symbols used for the choke control or tell-tale which were used on vehicles sold in the same international markets. Surveys of drivers at that time showed that what was understood from these symbols differed from country to country. Whilst a symbol was almost universally recognized in its country of origin, it was found to be unknown in others. It should therefore be apparent that population expectations of what a display means are based on prior experience and can be powerful influences on the success of a visual display. Clearly some form of standardization was required. The major developments in vehicle visual displays from 1970 to the present day lie in the area of display technology. International legislation has begun to harmonize the presentation of symbolic information for 216
Fig. 2 Instrument panel o f a 1968 Fiat 128
Fig. 3 Instrument panel o f a 1967 Honda $800
a
b
I ,1
e
Fig. 4 Three symbols o f choke control and displays; a - - UK; b - - France; c - - F R G (ISO 2575)
DISPLAYS. OCTOBER 1984
In recent years the development of many forms of electronic display technology has made a range of components available for use in vehicles. Initially these were small discrete displays dedicated to supplying supplementary information as an additional component to be mounted alongside conventional electromechanical instrument packages. Commonly these devices gave a digital display of a variable, the most common application being a clock. Further developments in display technology have now reached a point where complete instrument panels are produced in solid state electronics. Figure 5 shows a current production vehicle, an MG Maestro, fitted with such a panel. This current application of electronic displays therefore uses new technology to replace earlier electromechanically displayed functions on an item-for-item basis. Specific areas of these electronic instrument panels are uniquely used for displaying a specific function. However, the choice of format for this information display is made far wider than with conventional instruments. What then for the future? Currently, the implementation of electronic displays has affected only a small proportion of all new vehicles. Solid state electronic instrument panels have been fitted to vehicles at the higher price end of the market, or as top of the range options to more common models. However it is virtually certain that within a short time these displays will be fitted to the majority of new vehicles. After all the attractions to the designer and manufacturer are understandable. Electronic displays will integrate well with the increasing importance and frequency of the electronic management of a number of vehicle systems. Such system control functions will also provide the sensing and processing required for information display and could also provide the processing back-up for even more flexible presentation of that information. These electronic systems as a whole therefore offer advantages over conventional electromechanical systems in efficiency, potentially reduced weight and size, as well as increased flexibility. Also as research and development continue into the components themselves, increased reliability and production efficiencies will help to highlight the economic advantages of solid state electronics. In short, the electronic display will soon become the norm rather than the exception.
ERGONOMIC DESIGN What of the driver? He or she will clearly be affected by this potentially radical change in the means by which he is kept informed of vehicle system performance. It is obviously the role of the human factors specialist to assess the implications of this imminent change in vehicle design and to provide information on how this interface between driver and display, may be most successfully designed. This ergonomic assessment can be said to be at two levels, the general and the specific. First, the presentation of information to the driver should be re-evaluated in the light of new display technology. The driver's use of display should be assessed in the context of current vehicle and environmental conditions, and those likely in the future, to ensure that all aspects of the complex interaction between man and the machine are examined. Second, individual factors which are important to the total system need to be identified and classified. Specific problem areas can then be assessed under controlled conditions. In this manner the relative importance of these potential problems can be evaluated, and suitable solutions determined. A n a l y s i s o f the s y s t e m The visual display has been described as an interface between man and machine, driver and vehicle. This terminology is used to describe a site or location where information is passed between two components of a system. In present discussion these are two distinct aspects that affect the efficiency of this information transfer. These may be termed visibility and legibility. A display that is visible may be seen correctly by the viewer. A display that is legible however is also correctly understood by the viewer. As an example, an English book read by a non-English reader may be visible to him but will not be legible. However, a display that is not visible is hardly likely to be legible! These two aspects of the efficiency of the driver-display interface form the basis of the system analysis and therefore are critical to display design. There are four groups of factors which interact with one another to affect this success: they can be categorized as human, environmental, display and task factors. These are portrayed diagrammatically in Fig. 6. This analysis of the important factors contributing to the effectiveness of the visual display interface should not be though to be exhaustive, but should be viewed as a simplified model outlining the most crucial factors. A brief discussion of these individual factors is given in Tables I to 4. Specific ergonomic studies Following the analysis of the system which is described above, specific areas of factor interaction may be identified which are though likely adversely to affect display visibility or legibility. These potential problem areas are best examined in detail under controlled conditions.
Fig. 5 Instrument panel o f a 1983 MG Maestro
DISPLAYS. OCTOBER 1984
Recent research work performed at MIRA was undertaken to study several specific areas of interaction of these factors under laboratory conditions. This study was made possible by the construction of a vehicle 217
/
• • • •
Environment Illuminationleveland spectrum Vibration Spaciallocationof driver and display Glare SOurces
Human factors s Aoe/se~ = Visual acuity q i Colourvision • Fatigue • Expectancy • Night-timevisior • Language • Reaction speed • • • • •
[ Displayfactors [ I• Characterangularsize I J• Characterluminance/ J contrast I J• Character generationJ la Font/graphic design I J • Colour J
Environmental
factors
I Visual display J Visibility/legibility I
Task / Levelof workload Forward or reverse vehicle motion Display obscuration Type of information required Vehicle speed
Fig. 6 Driver.display interface Nctors simulator, designed to be able to vary these levels of factors that were identified as important for study. In particular, the following factors are controlled within the simulator.
•
Ambient illumination measured at the instrument panel, variable between 10--80 000 Ix; Correct spectral balance for chosen illumination level; Range of driver/display spatial locations to represent all road vehicle values; Facility to present glare sources.
Human factors • Adjustment to suit the range of sizes in the drivers population; Facility to measure the performance of the subjects in reading displays; Display factors • Ability to house a wide range of size and types of displays, with a variety of display characteristics; Facility to measure the photometric performance of each display; Task factors
• Facility to incorporate secondary tasks within the simulator.
The simulator was constructed to these parameters and ~ncorporated a total 1.35 kW of artificial illumination
Table 1. Human factors
Human Factor
Description
Age
The driving population ranges in age from 17 to more than 90 years old. Many aspects of human ability suffer impairment with increased age. Visual performance also follows this trend. Visual acuity and colour sensitivity are two examples, and their age effects are discussed below. Accommodation, or the ability of the eyes to focus at different distances, is affected by ageing. The nearest focusing distance for the normal 16-year old can be 80 ram. This may increase to 250 mm at 45 and to I metre at 60. RESULT: In specifying character size for displays used by the full range of drivers in the population, particular note must be taken of those over 60 years both in terms of acuity and accommodation.
Sex
The main characteristic of importance which is related to sex is colour vision occur mainly in the male population. RESULT: See colour vision below.
Visual acuity
Visual acuity can be described as the ability of the eye to resolve fine detail. The effect on performance depends on the size of target and its brightness. Consequently the level of illumination present and the contrast between target and background are important contributory factors in determining visual performance. This ability is age-dependent, ie with increased age visual acuity decreases. Some sources suggest performance decrements of 50 per cent for the population of 40+ compared to the 20-40 age group. Therefore drivers over 40 may require up to twice the character size as those younger at threshold values. Visual acuity is also impaired under night time conditions. RESULT: Again the age effects of acuity require a close consideration of the needs of the older driver.
Colour vision
Deficiencies in the colour discrimination ability of the population occur particularly in the male population. There are many forms of colour vision deficiency but the most common are difficulties in discriminating between red and green or blue and yellow. In addition colour discrimination generally deteriorates with increased age. RESULT: Once more the requirements of a portion of the population dictate the requirements for the overall population. Colour combinations to avoid are red/green and blue/yellow due to confusion problems.
218
DISPLAYS. OCTOBER 1984
mounted on multi-way adjustable frames within the simulator to give the desired levels and type of illumination. The simulator itself was constructed as a framework which represented a vehicle compartment with front bulkhead, instrument panel, driver's door and roof. In addition a steering wheel and column could be added. The driver's seat was adjustable in horizontal, lateral and vertical directions relative to the instrument panel which was also adjustable for angle of rake and lateral position. The simulator was housed in a room blacked out to external daylight. Three specific areas were studied using this simulator - - a m b i e n t illumination and glare; character size and
angle of view; and analogue or digital presentation of information; In all of these studies, small groups of subjects selected from the driving population were trained in the experimental tasks before their performance in using the display was measured. These performace measures did vary between studies but speed and accuracy of reading the display were the most important experimental response measures. These were chosen as they had been shown to provide a sensitive measure of the factors under study. Speed of reading was measured
Fig. 7 Vehicle simulator test rig by controlling the exposure of test displays to the subjects by means of a mechanical shutter. The simulator in this form is shown in Fig. 7. The simulator also has the facility to provide a secondary tracking task which is
Human Factor
Description
Fatigue
Fatigue in the context of driving is an important factor to consider where the question of crucial event information is concerned. Fatigue effects upon the driver are shown, in increasing order of importance; in inattention to individual components of the driving task, erratic visual search behaviour, and finally short periods of unconsciousness or 'microsleep'. A display cannot eradicate these effects, of course, but crucial event information should attract attention in an attempt to counter the less severe fatigue effects. RESULT: All crucial event information should be 'attention' attracting to the driver.
Expectancy
The driver population as a whole has expectancies of the information and format of information given by vehicle displays. In terms of general format the common fixed scale-moving pointer display has been present in practically all vehicles for major display functions. Hence the driving population expect this format, and are trained to use it. Colour coding such as 'red' warning lights, 'blue' main headlamp beam etc are accepted and standardized. However symbology for functions differs between populations. RESULT: Novel display formats should be evaluated for their potential acceptability by target populations before use. Careful note of population expectations should be undertaken before design for any particular market.
Night-time vision
The eye uses two mechanisms of the retina to detect light. These are described by their functional units in the retina, the rods and cones. The cones are primarily sensitive to the wavelength of light and therefore colour, and are predominant around the optical axis of the eye. The rods are more sensitive to the amount of light present and are mainly situated at the periphery of vision. RESULT: Colour discrimination at the periphery of vision during night-time conditions is seriously impaired. Therefore crucial event information is required to be sufficiently attention seeking to attract the central gaze of the eye which is more sensitive to colour.
Language
This applies mainly to alphanumeric displays which could present information in a written form. Intelligibility should be assessed for all countries where the vehicle will be used. RESULT: Consider target markets in providing coded information.
Reaction speed
There is variation amongst the driving population in almost every respect, including response speed. In terms of visual display this requires consideration of the format and content of information so that it is easily assimilated by the slowest responders. RESULT: Evaluation of novel designs requires testing by a wide range of subjects from the population.
DISPLAYS. OCTOBER 1984
219
Table 2. Environmental factors
Environmental factors Illumination level and spectrum
Description The illumination level found at the instrument and spectrum panel surface of a passenger vehicle is of course variable with the time of day and season of the year. For European driving it may be estimated that the range of levels may be from 0.01 Ix (moonlight) to 60 000 Ix (direct side window solar illumination). The spectral qualities of the illumination present in a vehicle can be variable. Daylight conditions, where the sun is the light source, provide wide-band illumination even though this can take on pronounced casts depending upon solar altitude. However, more veiling illumination is provided by narrow-band emitting artificial road lights. RESULTS: High levels of direct illumination can cause wash-out of active displays, and colour of the incident light can greatly influence visibility, particularly of narrow-band emitting warning lights. Evaluation techniques for displays must accurately recreate both level and spectra of representation vehicle ambient illumination ranges, taking careful note of any specific operating conditions of special purpose vehicles.
Vibr~ion
Vibration induced in both the driver and any visual display in the instrument panel is created by complex interaction of the road surface, tyre and suspension characteristics. The result is a variable acceleration/freqency pattern during driving. Typical values noted for passenger care are 0.16 ms -z at 4 Hz and 0.1--0.3 ms -2 at 11 Hz. If a vehicle is due to operate under environments where more extreme vibration conditions may occur then of course more substantial effects may be seen. RESULTS: If vibration levels are likely to be severe then careful consideration of this factor is required perhaps to increase subtended angle of characters.
Spatial location of driver and display
The fascia/seating package of a vehicle imposes driver and display basic limitations on the display factor of subtended angle. For most normal passenger car packages a typical display viewing distance is 800 ram, and this has been used in this study. The angle of view of the display is also controlled by the manner in which the display is mounted in the fascia relative to the direct line of sight to the driver. RESULT: Current packaging suggests 800 mm as an acceptable viewing distance for the character sizes use in current visual displays, ideally placed so that the deviation required from the normal line of sight is minimized and that the displays are viewed normally.
Glare sources
These may be either natural sources (ie sun, either viewed directly or reflected) or artificial sources (vehicle headlights or roadside lights). Evidence from experiments suggests that current practice glare levels and display characteristics a not sufficiently near threshold values for this factor to be a problem. RESULT: If the level of glare is exceptional due to predicted operational conditions then further consideration should be given to display visibility factors.
analogous to that found while driving. This consisted of a pointer controlled by a steering wheel tracking a path drawn on a scroll pulled mechanically past a viewing aperture at windscreen level. Subjects were asked to read values displayed on their electronic instruments whilst performing the tracking task. A performance measure was derived in this case from the accuracy of both display reading and tracking ability. This arrangement is illustrated in Fig. 8. Computer-generated and controlled secondary tracking tasks are under development which can replace the mechanical arrangement used previously.
FUTURE DEVELOPMENTS Research data is available on many of the areas identified in the analysis given above. Indeed design guidelines have been developed from this information by a number of experts in this field. However, this should not suggest that all questions are answered and all problems solved. In an area of rapidly developing 220
technology some reassessment of current guidelines in the light of future innovations is required, quite apart from filling in those gaps that exist in our current knowledge. For these reasons laboratory-based research that uses facilities such as the MIRA vehicle simulator will continue to be required. This could then lead to a restatement of desirable design guidelines in the form of a specification based upon legibility test and performance criteria. This latter approach has been suggested by Gorrell e for assessing the legibility of video monitors and used a refined set of procedures for human factors testing. There is therefore a possibility that a similar approach could be developed for vehicle displays, which would undoubtedly be a more difficult undertaking due to the more complex nature of display use in a vehicle. However, there are also wider questions to be posed for the future presentation of information to the driver using electronic displays. If electronic displays are to be fitted to the majority of vehicles in the near future then it is pertinent to assess
DISPLAYS. OCTOBER 1984
Table 3. Task-related factors
Task-related factor
Description
Level of workload
The level of workload current on a driver during a journey is very variable. Quiet, relatively 'straight' rural roads offer relatively little workload in comparison to a congested town centre. In the context of visual displays this may seem that at times during a journey a driver has little or no spare capacity to view his instrument panel mounted display. RESULTS: The 'worst case' road situation which provides a high workload for the driver can completely interrupt the subsidiary task of periodically viewing the displays. If crucial event information (such as oil pressure failure) is to be conveyed to the driver successfully under these situations then it has to be of a form that can attract his attention, possibly by using a nonvisual mode.
Forward or reverse
This is the most obvious task-related factor to provide an interruption to the driver's ability to inspect his visual displays. When reversing a vehicle, the driver's view is normally directed rearwards and therefore away from the instrument panel. RESULT: As with 'level of workload' display reading tasks are potentially interrupted. Similar comments therefore apply. If crucial event information is to be given it must attract the driver's attention.
Type of information
A crucial factor is the appropriate determination required of type and format of information display. Various classifications have been advanced to describe the type of information required by a user from a display. Briefly, these can be described as quantitative, qualitative, check, and status/warning information. These are the principal visual display uses in a vehicle. Although with the future of electronic displays in mind, a fifth category which is the more detailed descriptive information provided by alphanumeric/symbolic displays may also be included. RESULT: In any assessment of a complete instrument panel design careful consideration should be given to each individual displayed item of information. What form of information is required from the display and how is this information coded to the driver in relation to all other displayed functions?
Vehicle speed
A special example of increased workload with higher vehicle speeds. Under motorway conditions, for example, the driver's visual search of the road is increasingly further ahead of the vehicle. Any distraction from this to inspect the instrument panel requires not only a substantial re-accommodation of the eye, but also causes a greater penalty in terms of road distance covered without the driver's view being directed to the road. RESULT: At high vehicle speeds the need for speed of reading from visual display is very much greater.
the radical departures from current practice that will be available to vehicle manufacturers and indicate which offers most promise for the driver. The current production status of electronic instrument panels has been described in a replication of earlier electromechanical implementations on an item-for-
Array of 3 fluorescent tubes
miler
'
~ ~/.~
/
/
Hopper collecting road pattern
Fig. 8 Vehicle simulator test rig
DISPLAYS. OCTOBER 1984
~ " ~ M a t t black roof section aperture ~ f o r tracking task ta! / Subject's ocular ~_~ . . . ~ . . ~ position
item basis. Therefore, while the medium and form for presenting information has changed, the basic task for the driver remains the same. He must still inspect the various displays on the instrument panel, process the information when received and then formulate and carry out an appropriate response. Specific human factors appraisals, sensibly constructed design guidelines and legibility tests may ensure that a full electronic instrument panel is at least as visible and legible as its earlier electromechanical counterparts. This situation does of course have the advantage that the driver's expectations of the form and functions of the instrument panel are not compromised. In short, to the driver the appearance of the instrument panel are not compromised. In short, to the driver the appearance of te instrument panel has changed little. However, this does mean that the visual inspection and information processing required by the driver remains largely the same. The increase in complexity of information presented on the instrument panel of vehicles has also been described. Part of the reason for this increase may be credited to the instrument panel's importance in
221
Table 4. Display factors Display factors
Description
Character angular size
The minimum acceptable size of a display character (normally expressed in minutes of arc) size is most closely related to the visual acuity of the driver and the location of the display. Taking note of character contrast, ambient illumination level, character font, human factors etc these have been accounted for. The minimum value suggested for accurate reading is: RESULT: 20 min of arc (ie 4 mm viewed at 800 mm)
Character luminance/contrast
In general terms, the legibility of any display is increased with greater contrast between symbol and background. In vehicle use this factor influences legibility as much as any other. Given many sources of research, and taking the minimum character angular size noted above, it is suggested that the value below should give acceptance legibility to the driving population. RESULT: At least 8:1; however, colour contrast is still in need of further research.
Character generation
For alphanumeric information in particular, two forms of character generation are available, segmented or dot matrix designs. Various studies have been performed in a variety of technologies attempting to assess the relative legibility of these formats. RESULT: The more segments present in a segmented display, ie seven segments and above then the greater the flexibility of character which may be presented. However, in most comparisons if contrast ratios a r e kept to acceptable levels (as given above) little difference is found between seven segment and various dot matrix characters (see below).
Font/graphic design
The graphic design or font of alphanumeric characters has been investigated at length by many researchers, much of the work referring to printed characters on conventional displays. This has evaluated many aspects of stroke-width and width-height ratios. RESULT: Many studies have shown remarkably small differences between fonts evaluated in dot matrix form: ASCII, Lincoln/MITRE and NAMEL are amongst those shown to be very similar in legibility tests. However, it is suggested that to display a range of non 'alphanumeric' symbols greater flexibility of character generation is needed than that available on a sevensegment character.
Colour
Colour vision deficiencies within the male population require that the use of colour within display designs in general should be controlled. There are, of course, existing legal requirements for colours used for displaying certain functions. Separating status and warning lights may be an important consideration, to provide position as well as colour coding of types of information along with any associated symbol, for particular vehicle functions. RESULT: It is important to consider display colour problems with particular reference to colour discrimination, combination of colours and symboloy.
Others
Current advances in electronics enable us to consider which sensory modalities should be used in specifying displays. Previous comments have suggested that crucial event displays (such as 'low oil pressure', 'overheating' and 'no fuel' warning require immediate action on the part of the driver. For a visual display to have the required attention attracting characteristics then briefly flashing 'emergency' lights or auditory signals could be used as a back-up to 'request' attention to the status/warning displays. RESULT: Modern electronics will enable a range of auditory messages to be presented to the driver either in symbolic (ie buzzer/bell etc) or verbal form to require attention to a particular vehicle function. The integration of audio and visual presentation of information could prove a significant improvement to the notification of vehicle system state by simplifying the visual load of the driver.
the interior trim of a vehicle, and therefore its use in establishing different models within a vehicle range. Generally speaking, the more expensive the model within a range, the greater the complexity of instrument panel information. However, the need for high current levels of information on display is questionable. Today's congested road conditions require the full attention of the driver to be directed towards the events in the external road environment in order that the journey may be completed safely. This may mean either 222
that the driver regularly inspects all displays provided and is interrupted in the driving task, or, at the other extreme, that he ignores the displayed information and concentrates solely on the immediate road and traffic conditions. Clearly the actual behaviour of a driver between those two extremes is controlled by other factors, principally the information processing load caused by road and traffic conditions. When traffic is extremely busy the driver will largely ignore displayed information and vice versa. A second consideration
DISPLAYS. OCTOBER 1984
which further confuses this discussion is that of the interpretation that the driver makes of this displayed information. Probably all drivers would grasp the significance of a fuel level instrument showing 'E', 'Empty' or '0'. However, how many drivers would have the technical understanding of vehicle functions to grasp the significance and possible consequences of a sudden drop in oil pressure? Secondly, how many drivers would then select an appropriate course of action to avoid the undesirable outcome. Perhaps it is more important then to use the flexibility of vehicle electronic systems and displays to lighten the information processing load of the driver, and to assist the technically inexperienced members of the population understand the information displayed. The use of dot-matrix forms of a number of display technologies has made possible a redefinable display panel which can produce both alphanumeric and graphical forms of information as desired. Prototype installations of this form of display have been shown in recent years fitted to experimental vehicles. The most advanced of these have used a single display panel that gives a simple display of a minimum number of functions, for example speed, fuel level and elapsed mileage only. However, the microprocessor controlling the display can be instructed by the driver to display more detailed information on specific functions when required, eg oil pressure, oil temperatures etc. This new information displayed on the single central display panel which after a short period returns to the original basic information. In addition these interactive processors perform individual system status checks on all monitored functions. Dependent upon the amount of processing ability these checks can either notify the driver of 'warning' conditions, ie safe limits exceeded, or 'caution' conditions, ie safe limits approaching. Clearly these systems could be developed to even give simple advice as messages when warning or caution conditions are reached. These would be obviously tailored to alleviate the problems of the non-technical drivers. It is now also feasible to present some of these alphanumeric messages as spoken messages. While there are obvious detailed problems of auditory legibility to be overcome, this dis-
DISPLAYS. OCTOBER 1984
Fig. 9 Cathode ray tube display play mode should not be neglected, and its integration with the visual displays used should be considered in detail. The future could therefore see a simplified presentation of constantly displayed information which on demand from the driver or internal diagnostic functions could present a wealth of detailed information if required. Simple messages would be able to advise the driver of critical system conditions or failures in such a way that they were universally understood. This however is some way in the future. On-board diagnostics and message centre form instrument panel displays are in their infancy for mass-produced vehicles. Figure 9 shows a CRT-based display illustrating how this approach may appear to the driver. It remains to be seen whether this is the likely appearance of future vehicle displays, and whether human factors guidelines are used to develop systems which are a help to the driver and not a hindrance.
References 1 ISO 2575 'Road v e h i c l e s - symbols for controls, indicators and tell-tales' 2 GorreH, E.L. 'A human engineering specification f o r legibility of alnumeric symbology on video monitor displays'DCIEM Tech Rep 80-R-26 (Defence and Civil Institute of Environmental Medicine, Depart ment of National Defence, Canada, 1980)
223