INTELLIGENT CONTROL FOR AN ACROBOT

The acrobot is an underactuated two-link planar robot that mimics the human acrobat who hangs from a bar and tries to swing up to a perfectl y balanced upside-down position with his/her hands still on the bar. I n this paper we develop intelligent controllers for swing-up and balan cing of the acrobot. In particular, we first develop classical, fuzzy, and adaptive fuzzy controllers to balance the acrobot in its inverted unstable equilibrium region. Next, a proportional-derivative (PD) con troller with inner-loop partial feedback linearization, a state-feedba ck, and a fuzzy controller are developed to swing up the acrobot from its stable equilibrium position to the inverted region, where we use a balancing controller to 'catch' and balance it. At the same time, we develop two genetic algorithms for tuning the balancing and swing-up c ontrollers, and show how these can be used to help optimize the perfor mance of the controllers. Overall, this paper provides (i) a case stud y of the development of a variety of intelligent controllers for a cha llenging application, (ii) a comparative analysis of intelligent vs. c onventional control methods (including the linear quadratic regulator and feedback linearization) for this application, and (iii) a case stu dy of the development of genetic algorithms for off-line computer-aide d-design of both conventional and intelligent control systems.