Generating High-Speed Dynamic Running Gaits in a Quadruped Robot Using an Evolutionary Search

Abstract

Over the past several decades, there has been a considerable interest in investigating high-speed dynamic gaits for legged robots. While much research has been published, both in the biomechanics and engineering fields regarding the analysis of these gaits, no single study has adequately characterized the dynamics of high-speed running as can be achieved in a realistic, yet simple, robotic system. The goal of this paper is to find the most energy-efficient, natural, and unconstrained gallop that can be achieved using a simulated quadrupedal robot with articulated legs, asymmetric mass distribution, and compliant legs. For comparison purposes, we also implement the bound and canter. The model used here is planar, although we will show that it captures much of the predominant dynamic characteristics observed in animals. While it is not our goal to prove anything about biological locomotion, the dynamic similarities between the gaits we produce and those found in animals does indicate a similar underlying dynamic mechanism. Thus, we will show that achieving natural, efficient high-speed locomotion is possible even with a fairly simple robotic system. To generate the high-speed gaits, we use an efficient evolutionary algorithm called set-based stochastic optimization. This algorithm finds open-loop control parameters to generate periodic trajectories for the body. Several alternative methods are tested to generate periodic trajectories for the legs. The combined solutions found by the evolutionary search and the periodic-leg methods, over a range of speeds up to 10.0 m/s, reveal "biological" characteristics that are emergent properties of the underlying gaits.

Biomechanical determinants of transverse and rotary gallop in cursorial mammals

“…The control of locomotion has been a great challenge in robotics, since the emergence of research into the development and production of legged robots (Fukuoka et al, 2003;Tsujita et al, 2009). Several simulation models have been developed, based on mathematical models used for tuning different parameters (Herr and McMahon, 2000;Herr and McMahon, 2001;Krasny and Orin, 2004;Marhefka et al, 2003). Galloping robots have recently been developed, demonstrating that small changes in control parameters can produce all combinations of biological gaits.…”

Locomotion skills for simulated quadrupeds

“…Marhefka et al [2003] develop a control strategy based on fuzzy logic to produce simulated planar gallops. Krasny and Orin [2004] use a search algorithm to explore the space of open-loop 2D gallops. Herr and McMahon [2001] develop and analyze feedforward and feedback strategies for the transverse gallop of a horse, as tested using a 10-link planar simulation.…”

Emergence of a quadrupedal bound gait as interaction among the brain, body and environment

“…This bound gait contains four phases (flight, forelegs stance, four legs stance and hindlegs stance). In case of clockwise transitions, the flight phase is called "extended flight phase" (Krasny et al 2004). In case of counterclockwise transitions, the flight phase is called "gathered flight phase".…”