Customer Reviews

Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches

5 star:		(11)
4 star:		(3)
3 star:		(0)
2 star:		(2)
1 star:		(1)

 

 

I agree with a previous reviewer in that out of all the books I have come across on Optimal Estimation, this is by far the most suitable for self-study. I have found his explanations to be concise and straight-forward. That is, he goes straight to the point and delivers the concepts using simple/common language which is non-characteristic for academic books in the areas of Systems and Control. For a sample text on the subject written by him check his article on Kalman Filtering on the site embedded[dot]com

While conducting research as part of an Independent Study course, I have treasured this book like no other since it continuously serves as a valuable reference. The first two chapters which review the underlying mathematics (linear algebra and probability) necessary for understanding the central themes of the book are also above the usual presentation in related books. Needless to say that readers should not expect to learn the Math from this book alone, however, they can expect to find in these chapters most of the topics that usually need a quick review to make sense of higher-level concepts in the text.

I cannot stress enough that his use of language and clear explanations make this an easy-to-read textbook which simplifies the understanding of the topics. Do not get me wrong though, to really understand the problem of state estimation the readers need to be quite prepared in different areas of Engineering and Mathematics (hence my motivation for self-study).

A very clear, well written book that takes you step by step from the algebra and statistics basics to the most advanced developments of dynamic systems. The first part of the book is about providing all the knowledge required for the rest of the book in linear system theory (1st chapter), probability theory (2nd chapter) and least square estimation (3rd chapter). These chapters are very clear and, in my opinion, easy to follow for the non specialist. The second part is about the core subject, Kalman filter. Again, it is very clear and the fact that it very consistent with the 1st part in term of notation makes it very readable. Subsequent parts are more advanced topics but again nicely elaborate on the previous chapters and hence very easy to understand. I'll repeat myself but that really what I enjoyed most with this book: it is very progressive and takes you step by step.
I even think this is the best technical book I have ever read. Dynamic systems made easy!

I have 4 books on Optimal state estimation:
_ Applied Optimal Estimation of Arthur Gelb.
_ Optimal Control and Estimation by Robert F. Stengel
_ Optimal Control and Estimation Theory by George M. Siouris
_ Optimal State Estimation By Dan Simon

Of the 4, Dan Simon's Optimal State Estimation is by far the most useful for a GNC Engineer like me. He strikes a good balance between theory and practice and his examples are really useful. I find his treatment of EKF excellent.

This book relates control theory elegantly, to those with a scientific background, but not much control theory history. Dan uses well laid out algorithmic approaches, suitable for programming, and examples to explain the details and show the complexities in action. I especially like the non-linear filtering chapters, and the comparison s between the Kalman Filter and other approaches (Particle Filter, etc.) I have several estimation/control theory texts, and this is the one I carry around with me.

Overall I found the book to be very well written with plenty of proofs, which I really like. In the sections that I looked at (with one exception) I did not find any errors. For the more recent material, towards the end, I would highly recommend looking at the original papers in addition to what's in the book. Out of necessity some things are glossed over. In the case of the unscented Kalman filter the algorithm presented in the book has some issues. It is not exactly the same algorithm as in the original papers and will in fact produce a covariance matrix which is incorrect. However, on the whole I would highly recommend the book, especially for self learners.

I found this book to be a very well written well organized tratment of optimal state estimation. It made the material and the logical connections - building from OLS & RLS - more accessible than any other text I have used. In parituclar, the integration of H_inf filtering and combined Kalman-H_inf filtering follows naturally while taking the reader to the edge of modern practice. In this well-conceived framework, the unscented KF and the particle filter are particularly well explained.

I found this much more accessible than Gelb, and (despite numerous typos which will undoubtedly be corrected in the second edition) recommend it highly as a basic text. The only other book I found so clear, logical and comprehensive is Ljung & Soderstrom's Theory and Practice of Recursive Idenfication, which is long out of print.

This is an excellent book to learn estimation theory. I found the introductory chapter very helpful. The chapters on EKF and UKF are quite good. Highly recommended.

I read this book to understand and study the Kalman-family filters. Although they are just a part of the contents of the book, I found the materials are organized very well and highly readable comparing to other books on similar topics. The author has a comprehensive understanding of these topics, and gave good illustrations of the backgrounds of these techniques, together with a solid references list. This is very valuable if you need grasp the whole idea behind each estimation method and to construct your own image of such topics. I recommend this book to anybody who is new to these topics and want to get such information in most efficient way.

Well, as a beginner in the area of tracking, this book may be good for you, due to its bottom-up writing approach. One of the best things of this book is that it covers some advanced and recent development, like UKF and particle filter, which cannot be easily found in other textbooks.

Good introductory book on the subject. It is very easy to understand and does not require much background from the reader. For the more technically advanced readers, the pace may be a bit too slow. But nevertheless, I think it's a good introductory/self-study book especially for engineers.

I would also recommend Gelb's book ` Applied Optimal Estimation `.