Researchers: Xiaoming Hu, in cooperation with Ruggero Frezza (Univ. of Padova) and Clyde F. Martin (Texas Tech. Univ.).
Sponsor: The Swedish Research Council for Engineering Sciences (TFR).
In recent years autonomous systems have attracted a great deal of attention worldwide. Despite some successful examples and recent progress, the theory still lags. The establishment of the research center and graduate school on autonomous systems at KTH, and the ECSEL project at Linköping University about four years ago, both funded by the Swedish foundation for strategic research, have contributed to the better understanding of autonomous systems. This project is related but complementary to the autonomous system project at KTH. Although we are motivated by autonomous systems, the subjects we study are also important, in their own right, to path planning, sensor fusion, nonlinear control and nonlinear observer and filter design.
An integral part in the design and operation of autonomous systems is path planning and following. Both are difficult problems in a realistic environment and for a realistic mobile system. Equally if not even more difficult , is the problem of observing the structure of a dynamic environment using state variables. In brief, in this project we will study how to use sensor data for sensing, modeling and control. Several relevant theoretical issues will be studied. They include: using splines and other methods for environment reconstruction and path planning, fusing data from different types of sensors for more robust environment reconstruction and state observation, robust path following control and global feedback stabilization. Our methods will be strongly motivated from approaches in systems and control theory.
Research 1999/2000
Semidefinite programming and structural optimization
Stochastic realization theory and identification