Abstract—Accurate location information has received
considerable attention, especially in unmanned aerial vehicle
(UAV) area. However, the traditional positioning system using
Global Positioning System (GPS) and Inertial Navigation
System(INS) fusion information remains unstable. This paper
introduces a new positioning system based on the ultra wideband
(UWB) technology, which can achieve a ranging accuracy of
centimeter level. We use a classical filter algorithm based on the
Kalman filter, which fusing the information provided by UWB
and INS to obtain an accurate location information. Simulation
results show that the proposed algorithm outperforms the fusion
of GPS and INS. UAV experiment results show that this
algorithm can provide an accurate and stable location
information.
I. INTRODUCTION
With the rapid development of the UAV, the accurate
location service has attracted extensive attention from
researchers. In recent years, INS and GPS are widely used in
UAV’s positioning and tracking. Based on the Kalman filter
algorithm, usually the inertial sensors can be fixed by the
positioning information provided by GPS. It has proven to be
stable to fuse the information provided by INS and GPS based
on the Kalman filter (KF).
However, it should be emphasized that the accuracy of
GPS positioning system is affected by multiple factors. For
example, the time synchronization signal error between base
stations is a fatal source of error. Another significant problem,
which cannot be neglected is the multi-path effect. Therefore,
the use of the GPS positioning system cannot achieve a high
positioning accuracy.
At present, the mainstream wireless positioning methods
on the market mainly include infrared positioning, ultrasonic
positioning, radio frequency identification positioning,
ultra-wideband positioning, WIFI positioning, Bluetooth
positioning and ZigBee positioning, etc. According to the
principle, we can simply divide them into four categories,
namely, Time of Arrival (TOA), Time Difference of Arrival
(TDOA), Angle of Arrival (AOA) and Radio Frequency
Zheng Zhang is with the School of Automation Science and Electrical
Engineering, Beijing University of Aeronautics and Astronautics, Beijing
100191 e-mail: zzhang_buaa@buaa.edu.cn).
Qingdong Li is with the School of Automation Science and Electrical
Engineering, Beijing University of Aeronautics and Astronautics, Beijing
100191 e-mail: liqingdong@buaa.edu.cn).
Xiwang Dong is with the School of Automation Science and Electrical
Engineering, Beijing University of Aeronautics and Astronautics, Beijing
100191 e-mail:xwdong@buaa.edu.cn).
Zhang Ren is with the School of Automation Science and Electrical
Engineering, Beijing University of Aeronautics and Astronautics, Beijing
100191 e-mail: Renzhang@buaa.edu.cn).
Identification (RFID). These positioning methods have their
own advantages and disadvantages. Among them, the most
promising technology is UWB positioning technology, which
is based on TDOA. By labeling transmitting an ultra-wide
band signal to four base stations, the UWB system could
calculate the real-time position of the label through the ground
station computer. The UWB positioning system can achieve a
100-picosecond level synchronization and a centimeter-level
positioning accuracy.
The KF was first proposed for integrating INS
measurements with GPS. To improve positioning accuracy,
we use an UKF algorithm for integrating navigation grade INS
measurements with UWB. In order to achieve the best
performance, we need to establish the accurate dynamic
model and the stochastic information. The simulation result
shows that the fuse of UWB and INS resulted in a 20% (RMS)
improvements in location estimation. UAV experiment
showed that the proposed positioning system outperforms the
other positioning system[1-3].
The key contribution of this paper is that a new
high-precision positioning system is proposed, through this
system the information on UWB and INS are fully integrated
and achieved a centimeter-level position accuracy.
The rest of this paper is organized as follows. Section Ⅱ
mainly describes the state equation and the meaning of each
variable. Section Ⅲ introduces the employment of KF[4].
Section Ⅳ presents simulation results. Conclusions are in Ⅴ.
II. PRELIMINARIES
A. Each state variable in positioning system and state
equation
To establish the equation of state of UWB/INS integrated
navigation system, we need to determine the form and
dimension of each state variable in the state equation. In this
paper, the platform misalignment angle, velocity error,
position error, and zero offset of gyro and acceleration are
selected as state variables.
X(t)=[ ]
enu e n u xyz x y z
VVV H
δδδδεεε
∇∇∇
,
where
(t)X
is a system state vector with 13-dimension,
,,
enu
are UAV’s attitude rotation vector error from the
direction of east, north and up, respectively.
,
enu
VVV
δδδ
,
are the velocity errors,
ε
is gyro constant
drifts,
H
δ
is position error of height and
∇
is accelerometer
zero errors.
A New Positioning System Based on UWB and INS Fusion
Information
Zheng Zhang, Qingdong Li, Xiwang Dong and Zhang Ren, IEEE