An Expert System using Computer Algebra for Design of Nonlinear Control Systems

A prototype expert system for the treatment of stochastic control and nonlinear signal processing problems has been developed in a collaborative effort with INRIA. The system is written for the most part in MACSYMA and LISP. It accepts user input (models and design objectives) in symbolic form; it carries out the basic analysis of the user´s problem in symbolic form using MACSYMA; and it produces output in the form of automatically generated FORTRAN programs for the final numerical reduction of the problem. In this talk we shall describe a component of the system which supports the design analysis of nonlinear control systems. The methodology is based on the Hunt-Su-Meyer technique for exact or approximate linearization of nonlinear control systems using coordinate transformations based on the intrinsic differential geometric structure of the control system, i.e., its controllability properties and Kronecker indices. Sample sessions are presented to illustrate operation of the software. The status of the overall expert system and plans for its further development are described.