Simulation of a dynamic vertical jump

Abstract

The aim of this paper is to propose a bipedal structure able to follow high acceleration movements. The vertical jump of a human has been chosen as input (coming from experiments) to validate the controller design as it is one of the most complex motion. The study concerns the low level of the biped control that is to say the control design of one leg made of three rigid bodies, each of them moved by a pneumatic actuator. An analogy between a pneumatic actuator and a physiological muscle is first proposed. A dynamic model of the leg is then presented decoupling the dynamic effects of the skeletal (as interactions between segments) from the dynamic effects of the muscles involved. The controller is based on the nonlinear theory (taking into account the actuator and the mechanical models), it ensures a dynamic tracking of position and force. Its originality lays in the consideration of impedance behaviour at each joint during free and constrained tasks. It leads to asymptotically stable (Popov criteria) control laws which are continuous between contact and non-contact phases enabling real-time computations. The simulation results clearly show the tracking of position and forces during the whole jump cycle.