scispace - formally typeset
Search or ask a question
Book ChapterDOI

Fractional Order Extended Kalman Filter for Attitude Estimation

TL;DR: The Fractional Order Extended Kalman Filter (FKF) approach is designed for estimating attitude with the help of inertial sensors in the attitude heading and reference system architecture.
Abstract: Attitude estimation is one of the core frameworks for a vehicle navigating with the help of inertial sensors such as accelerometer, gyroscope and magnetometer. Measurements obtained by these sensors are fused together to obtain vehicle attitude in the form of roll, pitch and yaw angles. Several state estimation frameworks have been proposed in the literature of which the extended Kalman filter and the complementary filtering based schemes are most popular. In this paper, the Fractional Order Extended Kalman Filter (FKF) approach is designed for estimating attitude with the help of inertial sensors in the attitude heading and reference system architecture. The FKF scheme is applied on the sensor data captured from commercial navigation units and compared with reference attitude for analysis. The simulations are carried out for varying fractional orders of different states and the corresponding results depict the dependency of estimation accuracy on system order.
Citations
More filters
Proceedings ArticleDOI
02 Dec 2022
TL;DR: In this paper , different configurations of analog filters are tested in order to possibly find a better solution than that offered by the algorithm, used in many low-cost applications of Unmanned Aerial Vehicle positional awareness.
Abstract: This work deals with the comparison between different analog implementations of complementary filters with a digital processing algorithm. For this purpose, different configurations of analog filters are tested in order to possibly find a better solution than that offered by the algorithm, used in many low-cost applications of Unmanned Aerial Vehicle positional awareness.
References
More filters
Journal ArticleDOI
TL;DR: The proposed fractional order PI controller (FOPI) with SLAM method is used in the simulation of navigation of NAO humanoid robot from Aldebaran and shows that the FOPI controller can reduce the error between the real position and estimated position.
Abstract: We present a fractional order PI controller (FOPI) with SLAM method, and the proposed method is used in the simulation of navigation of NAO humanoid robot from Aldebaran. We can discretize the transfer function by the Al-Alaoui generating function and then get the FOPI controller by Power Series Expansion (PSE). FOPI can be used as a correction part to reduce the accumulated error of SLAM. In the FOPI controller, the parameters () need to be tuned to obtain the best performance. Finally, we compare the results of position without controller and with PI controller, FOPI controller. The simulations show that the FOPI controller can reduce the error between the real position and estimated position. The proposed method is efficient and reliable for NAO navigation.

10 citations

Proceedings Article
09 Jul 2012
TL;DR: A novel INS/GPS fusion architecture is proposed that demonstrated a significant improvement in performance over the conventional KF based schemes, in tests done on realistic simulated aircraft data.
Abstract: In this paper, we address the issue of aircraft navigational state estimation from the perspective of (i) aircraft attitude estimation, also called as the attitude heading reference system (AHRS), and (ii) estimating the full inertial solution of the aircraft (position, velocity & attitude), also known as inertial navigation system-global positioning system (INS/GPS) fusion, in the presence of accelerometer and gyroscopic bias. A suite of nonlinear filters; two Kalman filter (KF) based — extended and unscented Kalman filter (EKF, UKF) and a non-KF based filter that is the nonlinear complementary filter (NCF) on the μμ(3) group, are studied and evaluated for the AHRS. In this paper we propose a novel INS/GPS fusion architecture that demonstrated a significant improvement in performance over the conventional KF based schemes, in tests done on realistic simulated aircraft data. In the proposed architecture, the attitude estimation is decoupled from the position and velocity estimation, by exploiting the NCF as it is known for its superior attitude and gyroscopic bias estimation performance. The position and velocity estimation is carried out by a conventional EKF. The crucial difference between KF based schemes and the NCF for attitude estimation is in the generation of the measurement set, which involves trigonometric inverses and are susceptible to singularities for KF based schemes, which the NCF avoids. Furthermore, the NCF algorithm is faster and computationally more efficient than a KF algorithm scheme since the NCF does not involve the computation of matrix inverses like KF based schemes.

10 citations

Book ChapterDOI
15 Aug 2017

10 citations

Journal ArticleDOI
29 Mar 2016
TL;DR: In this article, a human opinion dynamics (HOD)-based optimization technique and modifying the technique using maximum likelihood estimators were used to tune the EKF parameters for attitude estimation using Global Positioning System aided inertial sensors.
Abstract: Purpose – The purpose of this paper is to solve the problem of tuning of EKF parameters (process and measurement noise co-variance matrices) designed for attitude estimation using Global Positioning System (GPS) aided inertial sensors by employing a Human Opinion Dynamics (HOD)-based optimization technique and modifying the technique using maximum likelihood estimators and study its performance as compared to Particle Swarm Optimization (PSO) and manual tuning. Design/methodology/approach – A model for the determination of attitude of flight vehicles using inertial sensors and GPS measurement is designed and experiments are carried out to collect raw sensor and reference data. An HOD-based model is utilized to estimate the optimized process and measurement noise co-variance matrix. Added to it, few modifications are proposed in the HOD model by utilizing maximum likelihood estimator and finally the results obtained by the proposed schemes analysed. Findings – Analysis of the results shows that utilization...

8 citations

Proceedings ArticleDOI
Xiaopeng Wu1, Yonghui Sun1, Zhinong Wei1, Guoqiang Sun1, Lei Liu1 
27 Jul 2016
TL;DR: In this paper, the state estimation problem for discrete nonlinear fractional order systems with non-Gaussian Levy noises is discussed, and a novel fractional-order extended Kalman filter (EKF) designing strategy is developed for discrete NOMA systems.
Abstract: In this paper, the state estimation problem is discussed for discrete nonlinear fractional order system with non-Gaussian Levy noises. By disposing the non-Gaussian Levy noises, the system state vector and measurement vector can be approximated directly, which could enable to deduce the new system noise covariance matrix and measurement noise covariance matrix, respectively. Based on the proposed approximating method, a novel fractional order extended Kalman filter (EKF) designing strategy is developed for discrete nonlinear fractional order system with non-Gaussian Levy noises, which is more general and effective than some existing results. Finally, some simulation results are provided to verify and illustrate the effectiveness of the obtain results.

3 citations