Experimental Comparison of Sensor Fusion Algorithms for Attitude Estimation

Cavallo, Alberto; Cirillo, Andrea; Cirillo, Pasquale; Maria, G. De; Falco, P.; Natale, Ciro; Pirozzi, Salvatore

doi:10.3182/20140824-6-za-1003.01173

Cited by 55 publications

(46 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…[1][2][3][4][5][6][7][8]). The classic approach consists in the use of a Kalman filter [9] and their nonlinear counterparts, such as the Extended Kalman Filter (EKF) or the Unscented Kalman Filter (UKF) [10,11].…”

Section: Introductionmentioning

confidence: 99%

Single axis FOG aided attitude estimation algorithm for mobile robots

et al. 2015

View full text Add to dashboard Cite

This paper is focused on an attitude estimation method for Autonomous Underwater Vehicles (AUVs).\ud Data acquired by a commercial Micro-Electro-Mechanical Systems (MEMS) Inertial Measurement Unit\ud (IMU), equipped with magnetometers, and a Fibre Optic Gyroscope (FOG) are fused to estimate the\ud attitude of the vehicle. One of the most used attitude estimation filter, a Nonlinear Complementary\ud Filter (NCF), is proposed as the basis of this work; then, some adaptations to the original formulation of\ud the filter are illustrated to better suit it to the field of underwater robotics. The proposed improvements\ud include the online tuning of the gains of the filter to cope with sensor disturbances and the employment\ud of the data acquired by a FOG. In addition, a fast procedure for the calibration of a magnetometer\ud is introduced to increase the reliability of its readings. The resulting filter is used to estimate the\ud attitude of an AUV; the performances of the proposed solution are tested and evaluated, in particular\ud when unpredictable magnetic disturbances are present, highlighting the improvements that the applied\ud changes allow to achieve in the specific field of application

show abstract

Section: Introductionmentioning

confidence: 99%

Single axis FOG aided attitude estimation algorithm for mobile robots

et al. 2015

View full text Add to dashboard Cite

show abstract

“…The task of analyzing the parameters of those three algorithms can be found in [2]. The EKF algorithm produces marginally better results regarding the attitude estimation accuracy but it requires higher execution times.…”

Section: Problem Formulationmentioning

confidence: 99%

An IMU-Based Wearable System for Automatic Pointing During Presentations

Sikeridis

Antonakopoulos

2016

Image Processing &Amp; Communications

View full text Add to dashboard Cite

show abstract

“…In this section, we will analyze the performance of three pose estimation fusion algorithms of IMU which include CF filter (proposed by Mahony), 10 Gradient Descent algorithm (proposed by Madgwick), 11 and EKF (proposed by Cavallo). 12 Mahony algorithm treats the pose estimation problem as a deterministic observation problem. The authors thought the definition of a Direct CF and a Passive CF arrive to a reformulation of the CF, named Explicit CF, in terms of direct vectorial measurements, such as gravitational or magnetic field directions obtained from an IMU.…”

Section: Imu Pose Estimationmentioning

confidence: 99%

“…And then we analyze three pose estimation fusion algorithms of IMU. By comparing with those methods which named CF algorithm, 10 gradient decrease algorithm, 11 and standard EKF fusion algorithm, 12 finally we take the standard EKF as the sensor fusion algorithm due to its good estimation result. Based on this, in this article, a new IMU-based iterative pose compensation algorithm is proposed based on the accuracy pose measurement of IMU.…”

Section: Introductionmentioning

confidence: 99%

“…The commonly used pose estimation algorithms are Complementary Filter (CF filter, proposed by Mahony), 10 Gradient Descent (proposed by Madgwick), 11 and Extended Kalman Filter (EKF) (proposed by Cavallo). 12 The CF algorithm doing low-pass filter on accelerometer measurement and doing high-pass filter after integrating measurement value of the gyroscope, and then, the pose angle is obtained by accumulating them together. Gradient descent algorithm, by using of magnetic field distortion compensation and using the quaternion representation, calculating data from the accelerometer and magnetometer for analysis and optimize the gradient descent algorithm.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Inertial measurement unit–based iterative pose compensation algorithm for low-cost modular manipulator

Lin

Min

Wei

2016

Advances in Mechanical Engineering

View full text Add to dashboard Cite

It is a necessary mean to realize the accurate motion control of the manipulator which uses end-effector pose correction method and compensation method. In this article, first, we established the kinematic model and error model of the modular manipulator (WUST-ARM), and then we discussed the measurement methods and precision of the inertial measurement unit sensor. The inertial measurement unit sensor is mounted on the end-effector of modular manipulator, to get the real-time pose of the end-effector. At last, a new inertial measurement unit-based iterative pose compensation algorithm is proposed. By applying this algorithm in the pose compensation experiment of modular manipulator which is composed of low-cost rotation joints, the results show that the inertial measurement unit can obtain a higher precision when in static state; it will accurately feedback to the control system with an accurate error compensation angle after a brief delay when the end-effector moves to the target point, and after compensation, the precision errors of roll angle, pitch angle, and yaw angle are reached at 0.05°, 0.01°, and 0.27°respectively. It proves that this low-cost method provides a new solution to improve the end-effector pose of low-cost modular manipulator.

show abstract

Experimental Comparison of Sensor Fusion Algorithms for Attitude Estimation

Cited by 55 publications

References 17 publications

Single axis FOG aided attitude estimation algorithm for mobile robots

Single axis FOG aided attitude estimation algorithm for mobile robots

An IMU-Based Wearable System for Automatic Pointing During Presentations

Inertial measurement unit–based iterative pose compensation algorithm for low-cost modular manipulator

Contact Info

Product

Resources

About