Orientation Estimation System Research Articles

A Robust Multi-State Constraint Optimization-Based Orientation Estimation System for Satcom-on-the-Move

Current orientation estimation system (OES) approaches can meet the Satcom-on-the-move (SOTM) accuracy requirement in low dynamics using inertial measurement unit (IMU) without Global Navigation Satellite System (GNSS). However, the estimation accuracy will decrease significantly due to the influence of the high dynamics of the vehicle. This article addresses this issue by constructing a multi-state constraint optimization-based OES (MSCO-OES). Our first contribution is a high-precision incremental inertial constraint, which adds the influence of the earth's rotation to the rotation part of the pre-integration theory to maintain the integral gyro accuracy for a longer time. The second contribution is that the nonlinear optimization method is applied to orientation estimation, making better use of historical information and having higher estimation accuracy and robustness through batch optimization and iterating. In addition, we use the lie group theory to extend the weighted least square estimation algorithm to the manifold, which can update multiple historical states to the current state in real-time through incremental inertial constraints. Compared with state-of-the-art methods, the proposed orientation estimation method has higher estimation precision and a reasonable calculation amount. The performance of the proposed method is demonstrated by comparing tests and SOTM deploying tests. Several high-dynamic and long-range tests show that the proposed OES has acceptable orientation estimation precision and reliability, which meets the high-reliable SOTM system requirements.

Robot Orientation Estimation Based on Single-Frame of Fish-eye Image

In order to develop the steering control for collision avoidance behaviour, robot must be able to determine its heading orientation with respect to environment. Orientation can be measured by dedicated sensors or through visual features perception. In vision-based orientation estimation problem, most of approaches are making use of a matching process between pair of frames. This paper proposes a method of estimating robot’s heading orientation by using only a single-frame of fish-eye image. CIE-LAB colour space is applied to handle colour and illumination intensity change. Straight line segments are extracted from thresholded CIE-LAB image take advantage of Progressive Probabilistic Hough Transform. Angle of the corresponding line segment is measured using combination of Law of Cosines and quadrant principle. Heading orientation in yaw angle is estimated by implementing voting mechanism based on region grouping and length of perpendicular line. Some experiments are made in robot soccer field environment to compare orientation estimation system against IMU’s measurement. Discussion about the performance and limitation of the system are included in this paper.

DEVELOPMENT OF HUMAN INERTIAL MOTION CAPTURE SYSTEM

This paper presents the full cycle of development of inertial motion capture system: from the selection of sensors, selection of the orientation estimation algorithm of the received data and system movement visualization. The paper considers the existing methods of motion capture using various physical principles of work. Inertial motion capture systems are used in the field of biomechanics increasingly. It is done the analysis of existing inertial motion capture systems in paper. There are several leading market manufacturers of inertial motion capture systems: Xsense, InertialLab, APDM. Comparative characteristics of these systems are given. It is proposed to use own inertial motion capture system (IMCS) as an alternative for expensive industrial designs.The study gives the concept of the developed system. The presented prototype of IMCS consists of autonomous AHRS-modules, base wireless receiver, subsystem of data collection and animation reproduction program. Each of the autonomous AHRS has its own power supply, sensors (IMU), microcontroller with firmware and radio transmitter. AHRS-module uses the ATmega328p microcontroller as calculation core. IMU (Inertial measurement unit) was created by combining on the same board the following sensors: gyro ITG3205 (InvenSense), accelerometer ADXL345L (AnalogDevices) and magnetometer HMC5883L (Honeywell). The firmware functional phases are described in paper. The first phase of the microcontroller firmware is reading data from sensors. After collecting the necessary information, the microcontroller program passes into the phase of the orientation estimation using the AHRS algorithm. The third firmware phase is the data packet preparation and its transmission to the receiver. In paper sensors’ calibration methods are described, the precision characteristics of AHRS-module are given. The developed IMCS has small dimensions of IMUs. They are autonomous and use wireless communication technology for data transfer.As AHRS algorithm of orientation estimation system uses complementary Bachman filter. It reduces the computational load on the microcontroller. Developed system has own software for object movement visualization. NI LabVIEW programming environment is used to display skeleton kinematics through data collected from AHRS modules. Shown approach uses method proposed by Denavit and Hartenberg to describe kinematic chain movement. It is used so called D-H parameters to get the equation of kinematic chain motion conversion. The developed IMCS can be configured flexibly for various kinds of tasks.

Performance Comparison of EKF-Based Algorithms for Orientation Estimation on Android Platform

Consumer electronics mobile devices, such as smartphones or tablets, are rapidly growing in computing power and are equipped with an increasing number of sensors. This enables to use a present-day mobile device as a viable platform for computation-intensive, real-time applications in navigation and guidance. In this paper, we present a study on the performance of the orientation estimation based on the data acquired by the accelerometer, magnetometer, and gyroscope in a mobile device. Reliable orientation estimation based on the readouts from inertial sensors may be used in more complex systems, e.g., to correct the orientation error of a visual odometry system. We present a rigorous derivation of the mathematical estimation model, and we thoroughly evaluate the performance of the orientation estimation mechanism available in the Android OS, and the proposed alternative solutions on an unique dataset gathered using an actual smartphone. From the experimental results, we draw the conclusions as to the best performing algorithm, and then we evaluate its execution time on Android-based devices to demonstrate the possibility of real-time usage. The Android code for the proposed orientation estimation system is made publicly available for scientific and commercial applications.

Design of orientation estimation system by inertial and magnetic sensors

The purpose of this paper is to present orientation estimation for small unmanned aerial vehicle (SUAV). An extended Kalman filter with adaptive PR (P denotes the estimation error covariance matrix, and R denotes the measurement noise covariance matrix) is designed to estimate orientation by sensors of gyroscope, accelerometer, and magnetometer integrated in Micro Electronic Mechanic System-based heading reference systems. Since ferromagnetic materials or other magnetic fields near the magnetometer disturb the measurement of local earth magnetic field and the external forces which produce maneuvering acceleration effect the measurement of gravity by the accelerometer, the orientation estimation is disturbed. Accordingly, the error equations of sensors are established using a current statistical model, and then the extended Kalman filter with adaptive PR with 12 state variables is designed. In the filter, the orientation error, gyroscope offset error, magnetic disturbance error, and maneuvering acceleration error are estimated. The swing experiment in hand with the magnetic disturbance and small maneuvering acceleration, and flight experiment for SUAV with the magnetic disturbance and large maneuvering acceleration, are developed. The compensation results show that the orientation is accurately calculated with disturbances. A new methodology for the orientation estimation is proposed, which could also be considered for other special application such as the robot on the ground and the autonomous underwater vehicles. This paper provides a novel realization method for accurate orientation estimation for SUAV. The method can be applied in many applications with a simple hardware.

Design and analysis of an orientation estimation system using coplanar gyro-free inertial measurement unit and magnetic sensors

Conventionally, there are two ways to determine the orientation angles of an object. One uses 3-axis accelerometers and 3-axis magnetic sensors, while the other one uses additional 3-axis gyroscopes to achieve a higher sensing accuracy than the first one. This paper presents a novel orientation estimation system that uses seven single-axis linear accelerometers and 3-axis magnetic sensors. These seven linear accelerometers are deployed in a way to form a “coplanar” gyro-free inertial measurement unit (IMU). As such, they can perform the integrated measurements of gravity force and angular velocity for the subsequent sensor fusion algorithms. With the information of the angular velocity, this novel design achieves a comparable performance with the one that includes 3-axis gyroscopes. The analysis of the system observability indicates that the angular velocity measurements can improve the observability of the angle estimation under certain circumstances. This result suggests the possibility of using fewer sensors in an orientation estimation system. Simulation results indicate that the proposed design achieve a sensing accuracy of 0.11°, 0.09°, and 0.2°for the roll, pitch, and yaw angle; this leads to the improvement of 540%, 632%, and 754% over the one that does not include the measurement of the angular velocity.

Aircraft recognition and orientation estimation using a structural skeleton model

Abstract In many vision applications, it is desirable to identify an object as a member of a class of objects rather than as a specific object in a class or to measure characteristics of the object without specifically identifying it. For example, one may wish to observe a fixed‐wing aircraft and make trajectory predictions without necessarily determining the specific type of aircraft. In this paper, a structural description for 2‐D views of aircraft is presented, a recognition technique is described, and orientation estimation is discussed. The structural description of the aircraft is based on skeletons of the aircraft images. The recognition and orientation estimation techniques rely on knowledge about the basic structure of aircraft. Although the initial algorithms are rather crude, they serve to establish the feasibility required to proceed with the development of a more advanced recognition and orientation estimation system.