IOT application in control system using wireless sensor fusion

Hong Duc University Journal of Science, E.3, Vol.8, P (138 - 144), 2017 
138 
IOT APPLICATION IN CONTROL SYSTEM USING WIRELESS 
SENSOR FUSION 
Pham Van Trung1 
Received: 15 March 2017 / Accepted: 7 June 2017 / Published: July 2017 
©Hong Duc University (HDU) and Hong Duc University Journal of Science 
Abstract: A wireless controller based on data fusion algorithm and IoT technology is 
designed and applied for an active QZS isolator. Firstly, an IoT sensor modular using 
wireless instead of wired connection is proposed. The wireless sensors of acceleration, 
velocity and distance are used in the active QZS isolator in order to eliminate unpredictable 
disturbance causing by wire of sensors. Secondly, the data fusion algorithm embedded in 
wireless sensor module is described to show how to combine acceleration, velocity and 
distance data into one kind of information. The fuser with advanced fusion technique will 
protect control system from a suddenly disabled sensor. Lastly, the system controller uses 
data fusion to compute control signal for producing a governing force of the isolator. The 
experiment result shows out 60% better eliminating vibration of the proposed controller. 
Keywords: IoT, wireless sensing, sensor fusion, active QZS isolator. 
1. Introduction 
Today, smart grid, smart homes, smart water networks, intelligent transportation are 
infrastructure systems that connect our world more than we ever thought possible. The 
common vision of such systems is usually associated with single concept, the internet of 
things (IoT), where through the use of sensors, the entire physical infrastructure is closely 
coupled with information and communication technologies; where intelligent monitoring and 
management can be achieved via the usage of networked embedded devices. 
A wireless sensor network (WSN) is a network formed by a large number of sensor 
nodes where each node is equipped with a sensor to detect physical phenomena such as light, 
heat, pressure, speed, etc. The sensors collecting data which are combined in one kind of 
information is called sensor fusion. The control systems using wireless sensors fusion have 
been found many benefits of not only reducing the monetary and time expenses associated 
with the installation of wire-based systems but also guarantee for some missing sensors. The 
use of wireless communications within a structural health monitoring data acquisition system 
was illustrated by Straser and Kiremidjian [1]. In addition, multi-sensor data fusion is the 
process of combining observations from a number of different sensors to provide a robust and 
Pham Van Trung 
Faculty of Engineering and Technology, Hong Duc University 
Email: Phamtrung85@gmail.com ( ) 
Hong Duc University Journal of Science, E.3, Vol.8, P (138 - 144), 2017 
139 
complete description of an environment or system states which is used in many areas of 
robotics such as such as object recognition, environment mapping, and localization [2]. 
In this paper, a wireless sensor fusion module is designed and applied for an active QZS 
isolator for some purposes: Firstly, using wireless sensor in the active QZS isolator is to 
eliminate unpredictable disturbance causing by connecting wires of the wired sensors for the 
isolator system with almost zero stiffness is easily effected by sensitive noise [3]. Secondly, 
the data fusion technique embedded in wireless sensor module is to protect control system 
from a suddenly disabled sensor while the system is running. Lastly, the system controller 
uses data fusion to compute control rule for producing a governing force of the isolator. 
2. Wireless sensor fusion module 
2.1. Hardware configuration of wireless sensor module 
The wireless sensor model includes two parts: Sensing and control interfaces as shown in 
Figure 1. The sensing interface is the Texas Instruments ADS834 AD converter which offers a 
16-bit conversion resolution and 4 sensing channels which is capable of digitizing any analog 
signal in the 0- 5 V range at sample rates as high as 100 kHz. The control interface is designed 
with a 16-bit digital - to - analog converter (Analog Devices AD5542) which receives binary 
number from the microcontroller and converts them to analog voltage signals. Both use the 
Atmel ATmega128 microcontroller for computational core where embedded software is stored 
to execute data acquisition and transition. Two WiFi/DSL routers, ENC28J60 Ethernet 
controllers are selected for wireless communication channel in both interfaces. 
Figure 1. Architecture of a wireless sensor interfaces 
2.2. Multi-sensor fusion based on Kalman filter 
In discrete-time controlled process, a Kalman filter is governed by the linear stochastic 
difference equations (1) and (2) which is described by an ongoing cycle as shown in Figure 2: 
Hong Duc University Journal of Science, E.3, Vol.8, P (138 - 144), 2017 
140 
1 1 1k k k kx Ax Bu w (1) 
k k kz Cx v (2) 
where A, B, and C are coefficient matrices; k is the time index; x is system states; u is 
control signal; z is measured states; w and v represent the process and measurement noise. 
Figure 2. Continuous cycle of Kalman filter 
A fusion technique is studied and embedded into the control interfaces of the wireless 
sensor module. In detail, displacement and velocity measurement signals are used for 
estimating acceleration, velocity and displacement signals based on Kalman filter algorithm. 
The outputs of the Kalman filters are data sources for a data fuser which will compute fused 
data providing for a system controller. The schematic model of sensor fusion is shown in the 
Figure 3 and the fusion algorithm is sketched in the Figure 4. 
Figure 3. Sensor fusion model 
Figure 4. Algorithm of data fusion 
Hong Duc University Journal of Science, E.3, Vol.8, P (138 - 144), 2017 
141 
3. Active control model of QZS isolator 
3.1. The QZS isolator using flexure 
Figure 5. QZS isolator using flexure [3] 
A mechanism of QZS isolator using flexure studied in this research is shown in the Fig. 
5. The mechanism consists of three main parts: the horizontal coil spring, vertical coil spring 
and notched flexures. While the vertical coil spring produces positive stiffness, the notched 
flexure under compressive force of initially deformed horizontal spring generates a negative 
stiffness which allows obtaining quasi-zero stiffness (QZS) characteristics. 
3.2. The motion equation of the isolator model 
Based on the dynamic analysis of this isolator model that is presented in the reference 
[3], the motion equation of dynamic model is derived as Eq. (3): 
3
2 ( )l n cmy cy k y k y f A y w t   (3) 
where y is vertical displacement; m is mass; c is system damping; kl and kn are linear 
and nonlinear stiffness respectively; fc is control force. 
3.3. Active control rule 
In studying the case of the horizontal actuation, the actuator force is derived based on 
the dynamic equation (3). The control law is derived and converted to the horizontal actuator 
force as shown in Eq. (4). 
3
1 2 3
0 0 0max( , ) max( , ) max( , )
nl
y y y
f a a a
y y y y y y

 (4) 
where a1, a2, a3 are Lyapunov control tuning gains. The a3 is gained for nonlinear 
feedback to cancel out the system nonlinear characteristic. The actuator force near equilibrium 
point (y=0) is considered to avoid saturation. 
Hong Duc University Journal of Science, E.3, Vol.8, P (138 - 144), 2017 
142 
4. Experiment configuration setup 
 The experiment configuration consists of passive QZS isolator, digital controller, 
sensors, actuator, and wireless sensor interface module as shown in Figure 6. Vertical passive 
isolator specification is summarized in the Table 1. 
Table 1. Isolator specifications 
Parameter Value Parameter Value 
m 25 - 40(Kg) Stroke ±0.005(m) 
khs 3.626x105(N/m) kvs 1.02x104(N/m) 
c 2.87(Ns/m) fn 1 (Hz) 
Several analog sensors are used for sensing system state data such as velocity sensor, 
acceleration sensor, and displacement sensor. The dSPACE controller computes and generates 
control signal from data provided by wireless sensor module. The control signal regulates the 
actuator through amplifier to stop vibration of the mass. 
Figure 6. Experimental setup 
5. Experiment result 
The experiment of the active QSZ isolator using wireless sensor fusion module is 
implemented to investigate the performance of wireless sensor fusion module and control 
response. The wireless fusion signal is verified by wired sensor signal and examined under 
suddenly disabled sensors as shown in Figure 7 and Figure 8. Control performance is 
obtained with two vibration isolation testing standards: impulse disturbance rejection and 
vibration transmissibility. The result of impulse disturbance rejection is shown in Figure 9 
and the result of the vibration transmissibility is shown in Figure 10. Both results show that 
the active QSZ isolator using wireless sensor fusion data has a good performance of 
vibration isolation. In the time domain, settling time is reduced 75 percent by active control 
system comparing to the passive isolator. In the frequency domain, the resonance magnitude 
is degraded about 60 percent. 
Hong Duc University Journal of Science, E.3, Vol.8, P (138 - 144), 2017 
143 
Figure 7. Signals verification 
Figure 8. Fusion signals under missing sensors 
Figure 9. Disturbance rejection [3] 
Hong Duc University Journal of Science, E.3, Vol.8, P (138 - 144), 2017 
144 
Figure 10. Transmissibility [3] 
6. Conclusion 
The active QZS isolator based on sensor fusion using IoT technology is proposed and 
investigated through the experiment. The proposed wireless sensor fusion module works well 
with data acquisition and system monitoring. The experiment results of active system 
responses show a good performance to vibration isolation in both time domain and frequency 
domain of the active QZS isolator by using fused data. 
References 
[1] E. G. Straser, A. S. Kiremidjian (1998), A modular, wireless damage monitoring system 
for structures, ReportNo.129, JohnA.Blume Earthquake Engineering Research Center, 
Department of Civil & Environmental Engineering, Stanford University, CA 1998. 
[2] M. Kam, X. Zhu, P. Kalata (1997), Sensor fusion for mobile robot navigation, IEEE 
Journals & Magazines, Vol. 85, pp. 108-119, Japan. 
[3] P. V. Trung, K. R. Kim, H. J. Ahn (2013), Nonlinear control of active QZS isolator 
based on lyapunov function, Korean Society for Precision Engineering and Springer-
Verlag Berlin Heidelberg, Volume 14, Issue 6, pp 919 -924, June 2013.