IEEE Consumer Electronics Magazine - September/October 2019 - 84

Special Section on Human Centric AR&VR Display and Human Interface Technologies for Automobile

Figure 4. Number of collisions recorded with HDD
versus HUD interface.

The participants were asked prior to the experiment, to follow the limits and driving rules set by
the British Highway Code.

EVALUATION RESULTS
The evaluation results presented encouraging data with regards to the collision occurrences with and without the HUD interface as
presented in Figure 4. The statistical significance of results was extrapolated from the
sample of the study (20 subjects) to the overall
population of drivers (large sample confidence
interval (CI) for the population mean). This
was calculated with the traditional large sample CI statistical analysis method. The collision
occurrences results were calculated with confidence of 95%, which suggests a margin of 5%
of potential error that is acceptable for the
nature of this evaluation. The analysis
highlighted that drivers have a probability of
an average 80% (0.99-0.61) collision when they
use the HDD as an infotainment source in normal driving conditions. This was decreased to
35% (0.58-0.12) when the proposed HUD interface was utilized.
This resulted in a sharp decrease of 45% of
collision's occurrences with the use of the prototype, gesture recognition HUD system. This
occurred due to the fact that the AR HUD interface in conjunction with the gesture recognition
interface enabled the drivers to maintain the eye
gaze on the road and their hands on the steering
wheel at all times. The drivers' speed perception
was also a revealing factor as 70% of the drivers

Figure 5. User Experience results: perceived task
load index of physical demand (left) and of mental
demand (right).

felt that they were driving faster and objectively
speed was marginally higher in the HUD trial.
Interestingly,
their collision occurrences
dropped nevertheless as it is highlighted in
Figure 4. The driver's collision avoidance performance was also reflected by their subjective
feedback, which is essential for any application
of consumer electronics. The ease of use
responses was further investigated on a Task
Load Index of the Physical Demand, which
offered overall a 65% of very low and low ratings
as illustrated in Figure 5 (left). Yet, 25% of the
users found that the physical demand required
was medium and 10% consider it high. The reason for this is that the seating position could not
facilitate perfectly all the users and as a result of
this, some of them found difficult the process of
successfully "air-clicking" the HUD icons.
A similar picture was presented on the Mental Demand workload with marginally better
results of 75% consisting of low and very low
responses in contrast to 25% of medium and
high as depicted in Figure 5 (right). The overall
consensus with regards to the system adoption
by the users instead of the traditional touchscreen HDD interfaces received a 75% acceptance, while 25% of the users were undecided.
Yet, the feedback informed the amendments of a
second version which the authors endeavor to
finalize and test in the near future.

CONCLUSION
This paper presented a prototype gesture
recognition HUD interface that was developed
with the aim to mitigate driver's distraction
occurring due to the plethora of incoming

IEEE Consumer Electronics Magazine

IEEE Consumer Electronics Magazine - September/October 2019

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - September/October 2019