Multi-sensor Data Fusion Using Adaptive Kalman Filter

2020 ◽  
pp. 2314-2320
Author(s):  
Yinjing Guo ◽  
Manlin Zhang ◽  
Fong Kang ◽  
Wenjian Yang ◽  
Yujie Zhou
Keyword(s):  
Kalman Filter ◽  
Data Fusion ◽  
Sensor Data ◽  
Adaptive Kalman Filter ◽  
Sensor Data Fusion ◽  
Multi Sensor Data Fusion

Weighted Multi-Sensor Data Fusion Based on Fuzzy Kalman Filter for Seam Tracking of the Welding Robots

2012 ◽  
Vol 542-543 ◽  
pp. 800-805 ◽  
Author(s):  
Jun Du ◽  
Mei Sun ◽  
Liang Hua ◽  
Jia Sheng Ge ◽  
Ju Ping Gu
Keyword(s):  
Kalman Filter ◽  
Data Fusion ◽  
Maximum Modulus ◽  
Seam Tracking ◽  
Measurement Noise ◽  
Sensor Data ◽  
Inference System ◽  
Sensor Data Fusion ◽  
Noise Characteristics ◽  
Multi Sensor Data Fusion

In order to resolve the problem of seam tracking of the welding robots with unknown noise characteristics, a Weighted Multi-Sensor Data Fusion (MSDF) algorithm based on the fuzzy Kalman filter algorithm is proposed. Firstly, each Fuzzy Kalman Filter (FKF) uses a fuzzy inference system based on a covariance matching technique to adjust the weight coefficient of measurement noise covariance matrix, so it makes measurement noise close to the true noise level. Secondly, a membership function in fuzzy set is used to measure the mutual support degree matrix of each FKF and corresponding weight coefficients are allocated by this matrix’s maximum modulus eigenvectors, hence, the final expression of data fusion is obtained. Finally, simulation results show that MSDF in seam tracking has both high precision and strong ability of stableness.

Multi-Sensor Data Fusion for a Tethered Unmanned Helicopter Using a Square-Root Unscented Kalman Filter

2016 ◽  
Vol 04 (04) ◽  
pp. 273-287
Author(s):  
Luis A. Sandino ◽  
Manuel Bejar ◽  
Konstantin Kondak ◽  
Anibal Ollero
Keyword(s):  
Kalman Filter ◽  
Data Fusion ◽  
Process Model ◽  
Position Estimation ◽  
Unmanned Aircraft ◽  
Sensor Data ◽  
Unmanned Helicopter ◽  
Sensor Data Fusion ◽  
Filter Performance ◽  
Multi Sensor Data Fusion

The use of tethered Unmanned Aircraft Systems (UAS) in aerial robotic applications is a relatively unexplored research field. This work addresses the attitude and position estimation of a small-size unmanned helicopter tethered to a moving platform using a multi-sensor data fusion algorithm based on a numerically efficient sigma-point Kalman filter implementation. For that purpose, the state prediction is performed using a kinematic process model driven by measurements of the inertial sensors (accelerometer and gyroscope) onboard the helicopter and the subsequent correction is done using information from additional sensors like magnetometer, barometric altimeter, LIDAR altimeter and magnetic encoders measuring the tether orientation relative to the helicopter. Assuming the tether is kept taut by an actuated device on the platform during the system operation, the helicopter position is estimated relative to the anchor point. Although this configuration avoids the need of a GPS, a standard operation mode for estimation of the absolute position (the position relative to the inertial reference frame) incorporating corrections with the GPS position and velocity measurements, is also implemented in order to highlight the benefits of the proposed tethered setup. The filter performance is evaluated in simulations.

Multi sensor data fusion based approach for the calibration of airdata systems

2011 ◽  
Vol 115 (1164) ◽  
pp. 113-122 ◽  
Author(s):  
M. Majeed ◽  
I. N. Kar
Keyword(s):  
Kalman Filter ◽  
Data Fusion ◽  
Test Data ◽  
Flight Control ◽  
Unscented Kalman Filter ◽  
Flight Test ◽  
Sensor Data ◽  
Sensor Data Fusion ◽  
Multi Sensor Data Fusion ◽  
Fusion Methods

AbstractAccurate and reliable airdata systems are critical for aircraft flight control system. In this paper, both extended Kalman filter (EKF) and unscented Kalman filter (UKF) based various multi sensor data fusion methods are applied to dynamic manoeuvres with rapid variations in the aircraft motion to calibrate the angle-of-attack (AOA) and angle-of-sideslip (AOSS) and are compared. The main goal of the investigations reported is to obtain online accurate flow angles from the measured vane deflection and differential pressures from probes sensitive to flow angles even in the adverse effect of wind or turbulence. The proposed algorithms are applied to both simulated as well as flight test data. Investigations are initially made using simulated flight data that include external winds and turbulence effects. When performance of the sensor fusion methods based on both EKF and UKF are compared, UKF is found to be better. The same procedures are then applied to flight test data of a high performance fighter aircraft. The results are verified with results obtained using proven an offline method, namely, output error method (OEM) for flight-path reconstruction (FPR) using ESTIMA software package. The consistently good results obtained using sensor data fusion approaches proposed in this paper establish that these approaches are of great value for online implementations.

Sign in / Sign up

Export Citation Format

Share Document