Page 213 - Autonomous Mobile Robots
P. 213

198                                    Autonomous Mobile Robots

                                starting point is completely different. First of all we show that, under some
                                technical assumptions, a nonholonomic system admits a smooth stabilizer only
                                if a subset of the differential equations describing the system are not defined
                                on a certain hyperplane passing through the origin of the coordinates system.
                                Hence, we focus on such a class of systems and we give sufficient conditions for
                                the existence of stabilizing control laws. Finally, we show that any smooth non-
                                holonomic system can be always transformed into a system which is not defined
                                on a certain hyperplane, say P, passing through the origin of the coordinates
                                system. Using the above results, we will propose in Section 5.3.4 a general
                                procedure to design discontinuous almost asymptotically stabilizers for non-
                                holonomic control systems. Such a procedure yields a control law which is not
                                defined on P; hence the closed loop system is not defined on P. However, we
                                will prove that every initial condition which lies outside P yields trajectories
                                which converge asymptotically to the origin.



                                5.3.1 Stabilization of Discontinuous Nonholonomic
                                       Systems

                                In this section we discuss the issue of smooth asymptotic stabilizability for
                                                                                 n
                                systems described by equations of the form (5.1) with x ∈ R and u ∈ R m+p .
                                We exploit a few basic facts from geometric control theory, as presented in
                                Reference 44. Note however that proper care has to be taken as we deal with
                                discontinuous functions.


                                Lemma 5.1   [20] Consider the system


                                                  ˙ x 1 = g 11 (x 1 , x 2 )u 1
                                                                                          (5.12)
                                                  ˙ x 2 = g 21 (x 1 , x 2 )u 1 + g 22 (x 1 , x 2 )u 2


                                            p
                                                                                             m
                                                                                    p
                                                                           n
                                with x 1 ∈ R , x 2 ∈ R n−p , x = col(x 1 , x 2 ) ∈ R , u 1 ∈ R , u 2 ∈ R ,
                                u = col(u 1 , u 2 ) ∈ R m+p , and m + p < n(g ij (x 1 , x 2 ) are matrix functions of
                                appropriate dimensions). Assume that the matrix function g 21 (x 1 , x 2 ) is smooth
                                in an open and dense set U, that the matrix functions g 11 (x 1 , x 2 ) and g 22 (x 1 , x 2 )
                                            ¯ 10
                                are smooth in U,  and that the distribution
                                                G = span{g 1 (x 1 , x 2 ), ... , g m+p (x 1 , x 2 )}

                                10  Let U be an open and dense set. We denote with ¯ U the smallest simply connected open set
                                properly containing U.




                                 © 2006 by Taylor & Francis Group, LLC



                                 FRANKL: “dk6033_c005” — 2006/3/31 — 16:42 — page 198 — #12
   208   209   210   211   212   213   214   215   216   217   218