Real-Time Implementation of GPS Aided Low-Cost Strapdown Inertial Navigation System

TLDR: This work details the study, development, and experimental implementation of GPS aided strapdown inertial navigation system (INS) using commercial off-the-shelf low-cost inertial measurement unit (IMU).

Abstract: This work details the study, development, and experimental implementation of GPS aided strapdown inertial navigation system (INS) using commercial off-the-shelf low-cost inertial measurement unit (IMU). The data provided by the inertial navigation mechanization is fused with GPS measurements using loosely-coupled linear Kalman filter implemented with the aid of MPC555 microcontroller. The accuracy of the estimation when utilizing a low-cost inertial navigation system (INS) is limited by the accuracy of the sensors used and the mathematical modeling of INS and the aiding sensors' errors. Therefore, the IMU data is fused with the GPS data to increase the accuracy of the integrated GPS/IMU system. The equations required for the local geographic frame mechanization are derived. The direction cosine matrix approach is selected to compute orientation angles and the unified mathematical framework is chosen for position/velocity algorithm computations. This selection resulted in significant reduction in mechanization errors. It is shown that the constructed GPS/IMU system is successfully implemented with an accurate and reliable performance.

Figures

Figure 6.16: GPS aided inertial navigation system Kalman filter algorithm data flow. The discretization time ∆t used in the algorithm is the sampling time of the inertial unit which is 0.02 seconds.

Figure 3.5: Side board of IMU which consists of rate-gyro sensor, this configuration will provide rotation around y-axis.

Figure 6.1: Strapdown inertial navigation system setup consist of 6 DOF IMU connected to PWM and Queued Analog to Digital Convertor (QADC) port. GPS is connected to the serial port of MPC555 Microcontroller.

Figure 6.5: Digital compass Honeywell Inc., HMR3000

Figure 3.10: Output Duty Cycle versus normalized specific force acting along accelerometer sensitivity y-axis.

Figure 3.9: Output Duty Cycle versus normalized specific force acting along accelerometer sensitivity x-axis.

Full text

Real-Time Implementation of GPS aided Low Cost Strapdown
Inertial Navigation System
A THESIS IN MECHATRONICS
Presented to the faculty of the American University of Sharjah
Collage of Engineering
in partial fulfillment of
the requirements for the degree of
MASTER OF SCIENCE
by
LAITH RASMI SAHAWNEH
B.S. 2002
Sharjah, UAE
April 2009

c
2009
LAITH RASMI SAHAWNEH
ALL RIGHTS RESERVED

Real-Time Implementation of GPS aided Low Cost Strapdown
Inertial Navigation System
Laith Rasmi Sahawneh, Candidate for Master of Science in Mechatronics
Engineering
American University of Sharjah, 2009
ABSTRACT
As autonomous navigation becomes considerable and important b e cause of
the increased demand to their usage and benefits, therefore reliability and integrity
issues become definite, specially when being implemented with commercially low-cost
sensors. The objective of this thesis is to both develop and implement in real-time
an INS/GPS integrated navigation system using the loosely-coupled linear Kalman
filter. The importance of the implemented algorithms are to function appropriately
and accurately using low cost inertial sensors where the rapid drift in sensors out-
put requires a reliance on external and available aiding source as Global Positioning
System (GPS).
This thesis describes the theoretical development and practical implementation
in real-time of strapdown inertial navigation system (INS) using commercial of the
shelf low cost inertial measurement unit (IMU) aided with the Global Positioning
System (GPS). When describing the IMU as a ”low cost”, this term means that this
unit is built using standard low grade accelerometers and gyros which cannot conduct
self-alignment. Therefore the thesis provides a desirable calibration procedure where
the requirements of a precisely controlled orientation of the IMU can be relaxed. To
do so, the thesis describes in detail the design, construction, error modelling analysis
and calibration approach of six-degree of freedom (6DOF) IMU. On the other hand,
calibration of (IMU) unit is one of the most challenging issues in navigation field
as it requires high accuracy measurements in order to maintain acceptable readings
from sensors, and normally the cost of calibration platform often exceeds the cost of
developing and constructing a MEMS sensor based IMU.
This thesis mainly discusses the development of inertial mechanization equa-
tions and algorithms that provides position, velocity and attitude of the host platform.
Then, the data provided by inertial navigation mechanization is fuse d with GPS mea-
surements using loosely-coupled linear. The accuracy of the estimation when utilizing
a low-cost inertial navigation system (INS) is limited by the accuracy of the used sen-
sors and the mathematical modelling of INS and the aiding sensors errors, however
iii

when fusing the INS data with GPS data, the errors can be bounded and the accuracy
will increase. This thesis provides, both in practical and theoretical terms, the fusion
processes adopted and real time implementation required for a high integrity aided
inertial navigation system using state-of- art technology microcontroller MPC555 as
the navigation computer. The theoretical work is verified by set of real-time experi-
ments using our developed INS/GPS equipment -The developed IMU, standard GPS
and MPC555-, the results have been compared with a top-notch INS/GPS Navigation
device. The experimental results using our designed INS/GPS has shown that posi-
tion and velocity accuracy can be archived using algorithms presented in this thesis
work.
iv

Contents
Abstract iii
List of Figures viii
List of Tables xiii
Nomenclature xiv
Acknowledgements xvii
1 Introduction 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Thesis Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.1 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2.2 Thesis Structure . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 Strapdown Inertial Navigation System 6
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1.1 Common Sensor Error Models . . . . . . . . . . . . . . . . . . 8
2.1.2 Initialization and Alignment . . . . . . . . . . . . . . . . . . . 9
2.1.3 Inertial Navigation System Error Models . . . . . . . . . . . . 9
2.2 Coordinate Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Equations of Navigation (overview) . . . . . . . . . . . . . . . . . . . 12
2.4 Strapdown Inertial Navigation Mechanization Equations . . . . . . . 14
2.4.1 Effect of Variation in the Earths Gravitational Field on The
Navigation Equations . . . . . . . . . . . . . . . . . . . . . . . 18
3 Design, Modelling and Calibration of A MEMS IMU 20
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.2 Design Specifications and Sensors Selection . . . . . . . . . . . . . . . 21
3.3 Sensors Error Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.4 Calibration Procedure And Experiment Setup . . . . . . . . . . . . . 34
3.5 Conclusions and Results . . . . . . . . . . . . . . . . . . . . . . . . . 39
v

Frequently asked questions

Q1. What have the authors contributed in "Real-time implementation of gps aided low cost strapdown inertial navigation system a thesis in mechatronics" ?

Therefore the thesis provides a desirable calibration procedure where the requirements of a precisely controlled orientation of the IMU can be relaxed. This thesis mainly discusses the development of inertial mechanization equations and algorithms that provides position, velocity and attitude of the host platform. Then, the data provided by inertial navigation mechanization is fused with GPS measurements using loosely-coupled linear. 

Q2. What are the future works mentioned in the paper "Real-time implementation of gps aided low cost strapdown inertial navigation system a thesis in mechatronics" ?

5 Conclusion and Future Work 119 using COTS components. 5 Conclusion and Future Work 120 the initial conditions of the desire variable of interest, GPS system and its most parameters, errors and aspects relating and affecting the fusion process has been discussed. This step was accomplished by using backlogged real-time data collected from both at the same time frame, the designed IMU ( AUSIMU ) /GPS, 6. 5 Conclusion and Future Work 121 IMU ( MIDG II ) from Microbotic, Inc. 6. 5. 2 Future Work 

Q3. What is the importance of the implemented algorithms?

The importance of the implemented algorithms are to function appropriately and accurately using low cost inertial sensors where the rapid drift in sensors output requires a reliance on external and available aiding source as Global Positioning System (GPS). 

Q4. What is the acceleration of an object produced by net forces?

Based on Newton’s second law of motion; the acceleration of an object is produced by net forces is directly proportional to the magnitude of the net forces acting on thebody, and inversely proportional to the mass of the object:fnet m = a = Fs (3.1)Where Fs is the specific force and a is the acceleration, which is independenton the mass. 

Q5. What is the purpose of this thesis?

This thesis mainly discusses the development of inertial mechanization equations and algorithms that provides position, velocity and attitude of the host platform. 

Q6. what is the s-function of the kalman filter?

The program model consists of I/O blocks, filtration, Kalman filter and data frame blocks which is an S-function blocks that can be programmed in C-code. . . . . . . . . . . . . . . . . . . . . . . . . . . 

Q7. Why is a calibration of an inertial instrument necessary?

Calibration of inertial instrument as discussed in the introduction of this chapter is necessary because the outputs are blend of accurate and erroneous. 

Q8. What is the calibration procedure for gyro sensors?

The calibration process is based on performing a series of manual rotations from −90◦ to +90◦ with a step of 5◦ along each accelerometer sensor sensitivity axis and a series of angular rotations starting from −150◦/sec to 150◦/sec with a step of 30◦/sec along each gyro sensor sensitivity axis. 

Q9. What is the purpose of the calibration procedure?

Modern calibration procedures utilize the benefits of Kalman filtering to obtain optimal estimates3.4 Calibration Procedure And Experiment Setup 35of the calibration coefficients. 

Q10. What is the objective of this thesis?

The objective of this thesis is to both develop and implement in real-time an INS/GPS integrated navigation system using the loosely-coupled linear Kalman filter. 

Q11. What is the MIDG II display and configuration program?

All MIDG II messages and configuration options are supported by the program.(Microbotic, Inc.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 6.8 MIDG II typical connection to a PC via RS232 port. 

Q12. Why is the angular rate of the gyro asymmetric?

The actual measured angular rate for each gyro can be given as:Vout = Sfω G oi + bG (3.21)Using the opposite sense then the authors can get:ωGoi = (Vout − bg)SfG (3.22)Where Vout is the gyro analog output voltage,ω G o is the angular rate actingalong the gyros input sensitive axis; i = x, y and z. 

Q13. What is the calibration coefficient for the gyro?

Then the calibration coefficients are evaluated [21]:βG = (X T GXG) −1XTGYG (3.26)A platform rotation schedule should be designed to provide measurement residuals that, together as a whole reflect the effect of all of the accelerometer and gyro calibration coefficients as shown in Figure 3.15.• 

Q14. Why is the accelerometer output output given in Figure 3.8?

The measured output of a single accelerometer as shown in Figure 3.8 can begiven as:Dout = Sfactor × g sin(θ) + ba (3.14)Where Dout is the accelerometer output duty cycle (refer to ADXL202EB data sheet), Sfactor is the scale factor, g is the gravity acceleration, g sin(θ) is the specific force along the accelerometer sensitive axis and ba is the accelerometer nonzero bias output, this output can be measured even though there is no component of specific force acting along the input axis. 

Q15. Why is the random walk a problem?

The integration of this random walk will result in velocity and positions moving at different rates during different runs even the IMU (and vehicle) are in the same direction and experiencing the same acceleration during each run [3].