Page 235 - Autonomous Mobile Robots
P. 235

220                                    Autonomous Mobile Robots

                                          n
                                             c
                                                                                          ∗
                                                 a i
                                W(x) =    i=2 i |x i | , with c i > 0 a i ∈{2, 3, ...}. Then there exists T > 0
                                                     ∗
                                such that for all T ∈ (0, T ) the controller u T := (u 1T , u 2T ) where 20


                                  u 1T =− g 1 x 1 − ρ(W)(cos((k + 1)T) −  sin((k + 1)T))
                                                                   2

                                        +   ρ sin((k + 1)T)
                                          2
                                                          α
                                                        W| (2ρ(W) + 2(g 1 x 1 + ρ(W) cos((k + 1)T))
                                  u 2T =− g 2 sign(L f 2  W)|L f 2
                                        × cos((k + 1)T) −  g 1 x 1 sin((k + 1)T))         (5.37)
                                with g 1 > 0, g 2 > 0, a > 0 and a sufficiently small  > 0, is a SP-AS controller
                                for the system (5.36) and the function
                                                                2
                                                                        2
                                    V T (k, x) = (g 1 x 1 + ρ(W) cos(kT)) + ρ(W) −  g 1 x 1 ρ(W) sin(kT)
                                                                                          (5.38)
                                is a (strict) SP-AS Lyapunov function for the closed loop system (5.36), (5.37).

                                   The control law is similar to the one proposed in Reference 50 and the
                                Lyapunov function is a modification of the one proposed in Reference 10.
                                The proposed result provides a discrete-time counterpart and to some extent a
                                generalization of Theorem 2 of Reference 10. Theorem 5.8 states that V T is a
                                strict SP-AS Lyapunov function for the closed loop system. It is well known that
                                the existence of a strict SP-AS negative Lyapunov function allows to address
                                the stabilization problem in the presence of disturbances.

                                                              ∗
                                                                                          ∗
                                Proposition 5.5 [42] There exist T > 0 such that for all T ∈ (0, T ) the
                                controller (5.37) is a SP-ISS controller for system (5.36) and the function (5.38)
                                is a SP-ISS Lyapunov function for the closed loop system (5.36), (5.37.)

                                5.8.2 An Example: A Car-Like Vehicle Revisited
                                In this section we apply the proposed result to the model of a car-like vehicle
                                introduced in Section 5.5.1.  Consider the model (5.30) and the coordinate
                                transformation [50,52]

                                               x 1 = x
                                                      3
                                               x 2 = sec (θ) tan(φ)
                                                                                          (5.39)
                                                        3
                                               x 3 = x sec (θ) tan(φ) − l tan(θ)
                                                        1 2
                                                             3
                                               x 4 = ly + x sec (θ) tan(φ) − lx tan(θ)
                                                        2
                                20  f 2 denotes the vector [0, 1, x 1 , ... ,  1  x n−2    ∂W  f
                                                        (n−2)! 1  ] and L f 2  W =  ∂x 2 .


                                 © 2006 by Taylor & Francis Group, LLC



                                FRANKL: “dk6033_c005” — 2006/3/31 — 16:42 — page 220 — #34
   230   231   232   233   234   235   236   237   238   239   240