Page 121 - Kỷ yếu hội thảo khoa học quốc tế - Ứng dụng công nghệ mới trong công trình xanh , lần thứ 8

104 TRƯỜNG ĐẠI HỌC SƯ PHẠM KỸ THUẬT - ĐẠI HỌC ĐÀ NẴNG

Research results indicate that, when dealing with
objects measuring 100x100mm at a distance of
2 meters, the sensor successfully detected objects in
47 out of 50 measurements, accounting for 94% of the
measurement time. Only 6% of the time, the sensor
failed to detect an object and solely measured the
distance to the rear wall. For objects larger than
200x200mm, the sensor consistently detected their
presence. These findings demonstrate the sensor's
Fig. 8. Block diagram of the warning device capability to detect pedestrians and vehicles within its
IV. IMPLEMENTATION AND RESULT scanning range in traffic scenarios.
B. Testing the detection of various types of
vehicles participating in traffic

Fig. 9. Design of the sensor device (on the left)
and the warning device (on the right)
A. Object detection test

Fig. 11. Traffic vehicle detection test
An experiment was conducted on a road in Hanoi.
The road is 7 meters wide and we conducted a test to
see if we could detect traffic on a lane with a width
equivalent to 3.5 meters. The sensor was installed at a
height of 3 meters above the ground and the distance
from the sensor to the middle of the lane was about
Fig. 10. Testing the object detection 4.4 meters. This means there is a blind spot about
capability of the sensor 40cm below the road surface, the height of the
A fairly simple test is performed to determine the motorbike is about 1 meter. Therefore, we configured
detection capabilities of the sensor. We mounted the the sensor device to send a warning signal when it
sensor on a stand and placed several objects in front detects a distance difference greater than 30cm.
of it at different locations, ultimately a wall. Our goal Warning devices are placed 5 meters away to
is to evaluate the sensor's ability to detect objects facilitate monitoring. For two vehicles moving close
within the scanning range. An Arduino sketch together, we will count them as one and the device
consists of 50 consecutive readings (taking place will only send one warning signal.
about 5 seconds) with intervals of 1 meter, 2 meters,
3 meters and 4 meters. Table V. Statistics of the number of vehicles
participating in traffic
Table IV. The ability to detect objects
at different positions Vehicles participating Number Number Detection
of
of
Size(mm) in traffic detections vehicles rate
(%)
100x100 200x200 400x400
Distance Walkers 14 20 70
1 meter 50 (100%) 50 (100%) 50 (100%) Motorbikes 73 78 94
2 meters 47 (94%) 50 (100%) 50 (100%) Cars 35 35 100
3 meters 0 50 (100%) 50 (100%)
4 meters 0 50 (100%) 50 (100%)
ISBN: 978-604-80-9122-4