Distributed Estimation

Prev: Optimum Receiver Design for Estimation over Wireless Links

Course Home

Next: Intro to Distributed Control

In this lecture, we will take a look at the fundamentals of distributed estimation. We will consider a random variable being observed by mutiple sensors. Under the assumptions of Gaussian noises and linear measurements, we will derive the weighted covariance combination of estimators. We will then touch upon the issues of distributed static sensor fusion. Towards the end, we will look at the problem of dynamic sensor fusion, i.e., distributing a Kalman filter so that multiple sensors can estimate a dynamic random variable.

Lecture Materials

• Lecture: Distributed Estimation

Reading

• Please refresh the material on state estimation covered in Week 4.

• Consensus algorithms will be covered in detail in the class next week. We will also touch upon one such algorithm in passing. For more details, you can read the following paper.

  • "Consensus Problems in Networks of Agents with Switching Topology and Time-Delays", Reza Olfati Saber and Richard M. Murray, IEEE T. Automatic Control, 49(9):1520-1533, 2004.

• Additional references are mentioned in the lecture notes. I particularly recommend references 11, 12 and 25.

  • "On Optimal Track-to-Track Fusion," K. C. Chang, R. K. Saha and Y. Bar-Shalom, IEEE Transactions on Aerospace and Electronic Systems, AES-33:1271-1276, 1997.
  • "Track Association and Track Fusion with Nondeterministic Target Dynamics", S. Mori, W. H. Barker, C. Y. Chong and K. C. Chang, IEEE Transactiocs on Aerospace and Electronic Systems, AES-38:659-668, 2002.
  • "Architectectures and Algorithms for Track Association and Fusion," IEEE Aerospace and Electronic Systems Magazine, 15:5-13, 2000.

Most of the cited papers are available using IEEE Xplore[1]. If you are unable to obtain any, please send me [2] a mail.