IEEE Consumer Electronics Magazine - May 2018 - 19

the viewing angle is more than ! 45c as the marker cannot
be sufficiently obtained.

For the obstacle detection function,
the average current is only about
80 mA, and the battery life is
up to 3.5 h.

poWeR coNsumpTioN
In this article, to assess the power consumption of the proposed glasses-type wearable device, an experiment is conducted to measure the average current and the battery life for
each function on a 400-mAh lithium polymer battery. The
experimental result is shown in Table 3. For the obstacle
detection function, the average current is only about 80 mA,
and the battery life is up to 3.5 h. The positioning function is
the highest power consumer, and it will reduce the battery life
dramatically because power consumption for continuously
transmitting image data captured from the microcamera
through wireless communication is relatively high. However,
unlike the obstacle detection function, users can open the
positioning function only when they need it, so the battery
life would be more than 2 h under normal usage conditions.

discussioN
An approach other than the proposed system is utilizing the
time-of-flight (ToF) cameras that use certain light sources
that emit light pulses that are bounced back by any object
within the field of view. By measuring the propagation time
between the sensor and the object, the corresponding distance
can be calculated. With the utilization of an active sensor, the
improvement window for increasing the power consumption
is through decreasing the intensity of the light source. However, this compromise will reduce the maximum operating
range and increase the possibility of the sensor experiencing
noise from ambient infrared signals. For instance, the ToF
camera for Project Tango is manufactured by pmdtechnologies based on the IRS1145C ToF camera from Infineon that
consumes 0.3 W and has a maximum detection range of 4 m.
A similar system based on the IRS1125C ToF camera (higher

100
80
60

Location A
Location B
Location C

40
20

15 0 -15 -30
Viewing Angle (°)

-45

-60

FIgurE 13. The effect of the viewing angle on the detection rate.

Detection Rate (%)

To evaluate the detection rate of the obstacle detection function, the experiment is conducted while varying the width of
square boards placed at head level as obstacles. (The height of
the boards is 20 cm.) For each square board, the experiment
records the detection rate with different walking speeds; i.e.,
slow (2 km/h), normal (3 km/h), and quick (4 km/h). Because
the ultrasonic sensors installed in three different directions are
the same model, the experiment mainly measures the detection rate in the forward direction. The system will alert users
by audio notification when the distance between the user and
obstacle is fewer than 1.5 m. The experimental result is shown
in Figure 14. The obstacle detection function recognizes about
80% of square boards longer than 30 cm when the user is
walking at a normal speed. Although the detection rate
declines to 70% when the user is walking at a faster speed, it
is acceptable because people with visual impairment will not
walk fast most of the time. Thus, the system could help them
avoid most obstacles located above the waistline.

Detection Rate (%)

THe deTecTioN RATe of THe
obsTAcle deTecTioN fuNcTioN

100
80
60
40
20
0

150 120 100 80 70 60 50
Square Board Length (cm)
Slow Walk

Normal Walk

Quick Walk

FIgurE 14. The detection rate of the proposed obstacle detection
on various distances.

Table 3. Power consumption of the proposed
system.
Function

Average Current

Battery Life

Obstacles detection

80.01 mA

3.5 h

Positioning

391.41 mA

0.71 h

Overall functions

426.17 mA

0.66 h

resolution and frame rate than the IRS1145C) has a maximum range of up to 6 m with an average power consumption
of 4.5 W. With reduced intensity, the system achieves a lower
power consumption of 1.65 W, yet the maximum detection
decreases by 2 m. From this instance, the achievable range of
the low-cost ToF-based systems is limited and has a tradeoff
against the power demand.

concLuSIon
A novel wearable locating system was presented based on a
custom visual marker, which is detectable at a maximum of
MAy 2018

IEEE Consumer Electronics Magazine

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - May 2018