FP8-4: Estimation and Sensor Information Fusion

Aalborg University Esbjerg

Instructor

Youmin Zhang

Department of Computer Science and Engineering

Aalborg University Esbjerg

Niels Bohrs Vej 8

6700 Esbjerg, Denmark

Phone: (+45) 7912 7741 Fax: (+45) 7912 7710

Email: ymzhang@cs.aaue.dk Homepage: http://www.cs.aue.auc.dk/~ymzhang/

Time and Location

• Lectures: Thursdays, 8:15-11:40am, Room B201.

Course Description

• http://www.cs.aue.auc.dk/research/lce/8smester/fp8-4.PDF

Course Materials

• Recommended textbooks
• [BLK01] Yaakov Bar-Shalom, X. Rong Li, Thiagalingam Kirubarajan, Estimation with Applications to Tracking and Navigation, Wiley-Interscience, June 8, 2001, ISBN: 047141655X.
• [HM04] David L. Hall, Sonya A. H. McMullen, Mathematical Techniques in Multisensor Data Fusion, Artech House Publishers; 2nd edition, March 1, 2004, ISBN: 1580533353.
• Optional textbooks / references
• [GA01] Mohinder S. Grewal, Angus P. Andrews, Kalman Filtering: Theory and Practice Using MATLAB, 2nd Edition, Wiley-Interscience, January 2001, ISBN: 0-471-39254-5.
• [HL01] David L. Hall, James Llinas, Handbook of Multisensor Data Fusion, CRC Press June 20, 2001, ISBN: 0849323797.

Communication

• Course web site: http://www.cs.aue.auc.dk/~ymzhang/courses/ESIF/index.html. Check it out regularly for lecture notes, schedule changes, and other course-related announcements.
• Emails may also be used for announcement, and posting questions and answers.

Tentative Schedules

Date/Location	Lecture Topics	Notes	Readings and References	Exercises
Thurs. 10/2 B201	Overview of Kalman filter o The continuous-time Kalman filter o o o o o The discrete-time Kalman filter o o o o The extended Kalman filter	Lecture 1	The discrete-time Kalman filter (DKF): BLK01: Sections 5.1-5.3, or GA01: Sections 4.1-4.2 Introduction to Kalman filter: Welch & Bishop (Paper; Slides) ; Maybeck (Ch1) Lecture Notes from IRS-7: Yang (slides); Ole (slides)	1. Practise the Example in Sect. 5.3 (BLK01, pp. 218-232) using DynaEst 2. Computer Applications Problem 5-2 (BLK01, pp. 265) Software: example_KF
Thurs. 17/2 B201	Variants (extensions) of Kalman filter o The continuous-time Kalman filter o The linearized and extended Kalman filter o o o o Multiple-Model (MM) based Kalman filters o o o o o	Lecture 2	The continuous-time Kalman filter (CKF): BLK01: Sections 9.1-9.2, or GA01: Sections 4.3 The linearized and extended Kalman filters (LKF & EKF): BLK01: Sections 10.1-10.3, or GA01: Sections 5.1-5.8 The Multiple-Model (MM) based Kalman filters BLK01: Sections 11.6, 11.8 Li: Survey on MM-based filters (paper; book chapter) Bak: Lecture Notes (AAU): (weblink)	1. Problem 9-10 (BLK01, pp. 369). 2. Problem 10-4 (BLK01, pp. 416). 3. Practise IMM estimator in Sect. 11.6.8 (BLK01, pp. 460-465) using DynaEst Software: example_EKF (file1, file2, file3)
Thurs. 24/2 B201	Nonlinear/Combined state & parameter estimation o The extended Kalman filter (EKF) o o o o The two-stage Kalman filter o o (TSKF) o o o The unscented Kalman filter (UKF) (optional) o o o o o o	Lecture 3	The Multiple-Model (MM) based Kalman filters (cont’d) BLK01: Sections 11.6, 11.8 Li: Survey on MM-based filters (paper; book chapter) Bak: Lecture Notes (AAU): (weblink) The extended Kalman filter (EKF) approach BLK01: Sections 11.9 The two-stage Kalman filter (TSKF) approach Papers: Friedland (1969); Keller (1997); Hsieh (1999); Wu et al (2000) The unscented Kalman filter (UKF) (not covered in the lecture) Wan & Merwe: weblink AAU ESIF course homepage: weblink	1. Problem 11-4 (BLK01, pp. 486). 2. Write a MATLAB script to implement the TSKF. 3. Can we use TSKF (instead of EKF) for estimating state and unknown parameter ξ (damping coefficient) in the second-order system example shown in the lecture? Why can and why cannot? Software: example_EKF (file1, file2, file3)
Thurs. 03/3 B201	Introduction to sensor information fusion o What is sensor information fusion? o Fusion models and architectures o	Lecture 4	HM04: Chapter 1; Bak: Lecture Notes (AAU): Chapters 1-3 (weblink)
Thurs. 17/3 B201	Sensor information fusion techniques o Fusion models and architectures o Techniques for sensor fusion o	Lecture 5	HM04: Chapter 11; Chapter 2 (optional).

Useful Links

· “Estimation and Sensor Information Fusion” course at AAU: Part1, Part2

· Publications in Information and Systems Lab., University of New Orleans (Prof. X. Rong Li)

§ Publications on Maneuvering Target Tracking Surveys

§ Publications on Multiple-Model Estimation with Variable Structure

§ Publications on Optimal Linear Estimation Fusion

§ Publications on Information Fusion

§ Publications on Fault Inference, Control Systems, Neural Networks

· A comprehensive web resources on Kalman Filter by University of North Carolina at Chapel Hill

---

---

This homepage is created and maintained by Youmin Zhang.