Rayleigh Quotient Iteration for a Total Least Squares Filter in Robot Navigation
Noisy sensor data must be filtered to obtain the best estimate of the robot position in robot navigation. The discrete Kalman filter, usually used for predicting and detecting signals in communication and control problems has become a common method for reducing the effect of uncertainty from the sensor data. However, due to the special domain of robot navigation, the Kalman approach is very limited. Here we propose the use of a Total Least Squares Filter which is solved efficiently by the Rayleigh quotient iteration method. This filter is very promising for very large amounts of data and from our experiments we can obtain more precise accuracy faster with cubic convergence than with the Kalman filter.
KeywordsMobile Robot Kalman Filter Iteration Method Extended Kalman Filter Robot Navigation
Unable to display preview. Download preview PDF.
- D. L. Boley and K. T. Sutherland. Recursive total least squares: An alternative to the discrete Kalman filter. Technical Report CS-TR-93-32, Department of Computer Science, University of Minnesota, April 1993.Google Scholar
- C. E. Davila. Efficient recursive total least squares algorithm for FIR adaptive filtering. IEEE Transactions on Signal Processing, 42(2):180–268, 1994.Google Scholar
- H. Durrant-White, E. Bell, and P. Avery. The design of a radar-based navigation system for large outdoor vehicles. In Proceedings of 1995 International Conference on Robotics and Automation, IEEE, pp. 764-769, 1995.Google Scholar
- B. N. Parlett and W. Kahan. On the convergence of a practical QR algorithm. In A. J. H. Morrell, editor, Information Processing 68, North-Holland, Amsterdam, 1:114–118, 1969.Google Scholar
- A. Ruhe. Computation of eigenvalues and eigenvectors. In V.A. Baker, editor, Sparse Matrix Techniques, Lecture Notes in Mathematics 572, Springer-Verlag, pp. 130-184, 1977.Google Scholar
- D. B. Szyld. A two-level iterative method for large sparse generalized eigenvalue calculations. Ph.D. thesis, Department of Mathematics, New York University, New York, 1983.Google Scholar
- S. Van Huffel and J. Vandewalle. The Total Least Squares Problem: Computational Aspects and Analysis, volume 9 of Frontiers in Applied Mathematics. SIAM, Philadelphia, 1991.Google Scholar
- T. Yang and M. Lin. Iterative total least squares filter in robot navigation. In Proceedings of IEEE 1997 International Conference on Acoustics, Speech, and Signal Processing (ICASSP-97), Munich, Germany, 1997.Google Scholar