CONTROL OF SMART EXERCISE MACHINES .1. PROBLEM FORMULATION AND NONADAPTIVE CONTROL

This is the first part of a two-part paper on the design of intelligen t controllers for a class of exercise machines. Tbe control objective is to cause the user to exercise in a manner which optimizes a criteri on related to the user's mechanical power. The optimal exercise strate gy is determined by a biomechanical behavior of the individual user, w hich is assumed to satisfy an affine force-velocity relationship depen dent on the body geometric configuration, Consequently, the control sc heme must simultaneously: 1) identify the user's biomechanical behavio r; 2) optimize the controller; and 3) stabilize the system to the esti mated optimal states, Moreover, to ensure that the exercise machine is safe to operate, the control system guarantees that the interaction b etween the exercise machine and the user is passive. In this first par t of the paper, we formulate the control problem and propose a control ler structure which satisfies the safety requirement and is capable of causing the user to execute an arbitrary exercise strategy if the use r's biomechanical behavior is known, The controller is of the form of a dynamic damper and can be implemented using only passive mechanical components. Part II of this paper is concerned with the self-optimizat ion problem, in which both the determination of the optimal exercise s trategy and the execution of that strategy, when the user's biomechani cal behavior is unknown, must be considered.