ã¹ã©ã€ãã£ã³ã°ã¢ãŒãå¶åŸ¡å ¥é
察象ã·ã¹ãã ãæ§é çäžç¢ºããããã€å Žåã«æå¹ãªããã¹ãå¶åŸ¡æ³ã ä»åã¯SISOã·ã¹ãã ã®ã¿åãäžããããMIMOã«ãæ¡åŒµå¯èœã 詳ããã¯J. E. Slotine, W. Li, âApplied Nonlinear Controlâ, 1991ãåç §ã ä»®å® æ¬¡ã®ã·ã¹ãã ãå¶åŸ¡ããããšããã $$x^{(n)} = f({\bf x}) + b({\bf x})u$$ ããã§ã${\bf x}=[x \ \dot x \cdots x^{(n-1)}]^T$ã¯ç¶æ ãã¯ãã«, $u$ã¯ã¹ã«ã©ãŒã®å ¥åã§ããã ãŸãã$f$ã¯äžç¢ºãããå«ãé¢æ°ã§ãããšããã å ·äœçã«ã¯ãæã ã¯æšå®å€$\hat f$ã®ã¿ãç¥ãããšãã§ãããã®ãšãããããã®é¢æ°ã¯ããæ¢ç¥ã®é¢æ°$F({\bf x})$ã«å¯Ÿã㊠$$\left|\hat f-f\right|\le F$$ ãæºãããšããã$b$ããŸã $$0 < b_{min} \le b \le b_{max}$$ ã®äžç¢ºãããæã€ãšããæã ãç¥ãããšã®ã§ããå€ã¯æšå®å€ $$\hat b = (b_{min} b_{max})^{1/2}$$ ã®ã¿ã§ãããšããã åé¡ ${\bf x}$ãæãŸããç¶æ ${\bf x}_d=[x_d \ \dot x_d \cdots x^{(n-1)}_d]$ã«è¿œåŸããã$u$ãèšèšããã ãã ããæãŸããç¶æ ${\bf x}_d$ã¯${\bf x}_d(0) = {\bf x}(0)$ãæºãããšãã (ãã¯ãã«ã«ãªä»®å®ãããã§ã¯èª¬æããªã) ã å°åº ãŸãã¯ç°¡åãªã·ã¹ãã 㧠簡åã®ãã$b$ã«ã€ããŠã®äžç¢ºãããäžæŠç¡èŠããŠ$b({\bf x}) = 1$ãšãã察象ã·ã¹ãã ã¯ç°¡åãªäºæ¬¡ç³» $$\ddot x = f({\bf x}) + u$$ ã§ãããšããã ãŸãããã以åŸã$\tilde{\bf x} := {\bf x}-{\bf x}_d$ãšå®çŸ©ããã ããããæ£æ°$\lambda$ãèšèšããããã§ãç¶æ ãã£ãŒããã㯠$$u = -\hat f({\bf x}) + \ddot x_d - \lambda \dot{\tilde{x}}$$ ãçæãããšãéã«ãŒã系㯠$$\ddot{\tilde{x}} = - \lambda \dot{\tilde{x}}$$ ãšèšç®ã§ãã$\dot{\tilde{x}}\rightarrow 0$ããªãã¡${\bf x}\rightarrow{\bf x}_d$ãæºããã ãã ãããã®çµæãæãç«ã€ã®ã¯ã$f$ãäžç¢ºãããå«ãŸãªã (ããªãã¡$\hat f=f$ãæãç«ã€) ãšãã§ãããæšå®å€ãçå€ãšäžèŽããªãå Žåã¯ãã®éãã§ãªãã ...