| This study facilitates the implementation of truly dynamic running by addressing both the control and design of a biped robot. An analysis of the mechanics of running is presented, followed by control strategies and design models that are motivated by the mechanics.;A tractable model of the mechanics of running is presented through principles, measures, and a time-distributed representation of the stance phase. The Steady-State Index is proposed as a characterization of biped running. It identifies and relates the parameters affecting running and has broad applicability to both biological and robotic systems.;We present a simple, heuristic control strategy that is applicable to articulated legged robots. The strategy is conducive to real-time, highly dynamic implementation and can accommodate diverse configurations and diverse actuation systems. The mechanics model motivates the control strategy and provides the basis for its scalability. The strategy consists of a simple set of three rules, where the key rule considers the leg-length upon liftoff. This rule offers a simple control for both steady-state and accelerated running. The strategy was verified in simulation, where it was tested extensively on both telescoping and articulated bipeds.;In support of the design process, analytical models are presented for two aspects of running: the duty factor (DF) of the gait and the stiffness value of the leg. For a given running speed, an optimal DF exists that minimizes the energy expenditure. Based on a model of the energetics, we present a formula for the optimal DF. This formula is validated by both human data and simulation results. In addition, a model is presented for the stiffness value of the leg as a function of the physical properties, speed, and DF. The Gait Resonance Point is proposed as a design target for compliant running. At this point, the gait matches the spring resonance and the stiffness value becomes independent of the DF. |
