event

Ph.D. Thesis Proposal: Kyuman Lee

Primary tabs

Ph.D. Thesis Proposal by

Kyuman Lee

(Advisor:  Professor Eric N. Johnson)

“Adaptive Filtering for Vision-Aided Inertial Navigation”

3:00 p.m. Wednesday, July 26, 2017
Montgomery Knight Building Room 317

Abstract: 
With the advent of unmanned aerial vehicles (UAVs), a major area of interest in the research field of UAVs has been vision-aided inertial navigation systems (V-INS). Many missions of UAVs—reconnaissance, damage assessment, exploration, and other guidance, navigation, and control (GNC) tasks—often demand V-INS in more operational environments such as indoors, hostilities, and disasters. In V-INS, inertial measurement unit (IMU) dead reckoning generates the dynamic models of UAVs, and vision sensors extract information about the surrounding environment and determine features or points of interest. With these sensors, the most widely used algorithm for estimating vehicle and feature states of V-INS is an extended Kalman filter (EKF). The design of the standard EKF does not inherently allow for time offsets between the timestamps of the IMU and vision data, and the necessary assumption of the EKF is Gaussian and white noise. In fact, sensor-related delays, correlations of noise, or outliers that arise in various realistic conditions are unknown parameters. A lack of compensation of unknown parameters leads to a serious impact on the accuracy of the navigation systems. To compensate for uncertainties of the parameters, we require modified versions of the estimator or the incorporation of other techniques into the filter.

The main purpose of this thesis is to develop adaptive and robust V-INS for UAVs, in particular, those for situations pertaining to such unknown parameters. First, to fuse measurements with unknown time delays, this study incorporates parameter estimation and constrained filtering into state estimation. In addition, we use machine-learning techniques (e.g., the kernel embedding of distributions) to handle unknown correlations and dependences of each noise source. Unfortunately, few researchers have treated correlated noise in V-INS in great detail. Finally, typicality and eccentricity data analysis (TEDA) detects the real-time outliers of the IMU and vision data, and variational approximation for Bayesian inference derives how to compute the optimal precision matrices of both propagation and measurement outliers. Results from both Monte Carlo simulation and flight testing validate the improved accuracy and reliability of V-INS employing these adaptive filtering frameworks.

Committee Members:
Prof. Eric N. Johnson (Advisor), School of Aerospace Engineering
Prof. Eric Feron, School of Aerospace Engineering
Prof. E. Glenn Lightsey, School of Aerospace Engineering
Prof. Marcus J. Holzinger, School of Aerospace Engineering
Prof. Byron Boots, School of Interactive Computing

Status

  • Workflow Status:Published
  • Created By:Margaret Ojala
  • Created:07/24/2017
  • Modified By:Margaret Ojala
  • Modified:07/24/2017