Theories of bipedal walking: an odyssey

520C.L. Vaughan / Journal of Biomechanics 36 (2003) 513–523Higher Centreu 0Neural Rhythm Generator•u = q(u)MotorSensoryCommandSignalT r (u) S e (ϕ,F g )(b)Musculo-Skeletal System•ϕ = p(ϕ,F g )(a)ϕEnvironmentF g(c)Fig. 10. (a) A model of the locomotor system links control and effect with coupled differential equations (Taga, 1995). Global entrainment of theneural system on the one hand, and the musculoskeletal system plus environment on the other, generates bipedal gait on (b) the level and can adapt to(c) an uphill pathway.coupled differential equations defined both the neuralrhythms (CPGs) and the mechanics of the musculoskeletalsystem (cf. Fig. 10a). Once the model had beentrained, it not only produced level gait under normalconditions (cf. Fig. 10b), but it also adapted to environmentalperturbations such as uneven terrain or increasedcarrying load (cf. Fig. 10c). Taga (1995) demonstratedthat the speed of walking could be controlled by a singleparameter which drove the neural oscillators, and thestep cycle could be entrained by a rhythmic input to theoscillators.7. Robots on two legsPowered dynamic gait in a bipedal robot can berealised only through a strategy which is based onstability and real-time feedback controlFive decades ago Isaac Asimov, the science fictionwriter (who also held a doctorate in chemistry)formulated his three laws of robotics. The second ofthese stated that ‘‘A robot must obey the orders given itby human beings’’. One of these orders would no doubthave been to walk, a natural extension of the originalfunction of a ‘‘robot’’, first introduced to the Englishlanguage by the playwright Karel Capek in 1921. Basedon his mother tongue of Czech, a robota is defined as aworker of forced labour. Over 30 years before Capekcoined the term robot, George Fallis in the USAinvented a bipedal walking toy (Fallis, 1888). Thecentral claim of his patent stated ‘‘This inventionconsists of a toy which is designed to simulate thehuman frame and which is a combined pendulum androcker construction, whereby when placed upon aninclined plane it will be caused by the force of its owngravity to automatically step out and walk down thesaid plane’’ (cf. Fig. 11).What Fallis had described was a passive bipedalrobot, where the word ‘‘passive’’ connotes the lack ofactive power. Another feature of the Fallis walker wasthat its gait was almost certainly static, in the sense thatits centre of gravity was always located within its base ofsupport. This is in contrast to dynamic walking wherethe centre of gravity falls outside the support baseduring the transition from one foot to the other. Thepassive vs. active debate is an interesting one, as is thediscourse on static vs. dynamic gait. A little over 10years ago McGeer (1990) demonstrated that a passivewalker based on the Fallis design, despite the lack of anyform of feedback control, could achieve dynamic gait.His work has led to more recent efforts to explore thepotential of passive gait to provide biomechanicalinsight (Garcia et al., 1998; Kuo, 2001). The firsttheoretical attempts to describe the mechanics of anactive bipedal robot were made over 30 years ago(McGhee, 1968; Vukobratovic, 1970; Frank, 1970). Agroup at Waseda University in Japan, led by Kato et al.(1974), are generally acknowledged to have been the firstto design and implement a successful active walker.However, it employed a static gait pattern, with activewalking in a bipedal robot only being achieved in thelate 1980s (Zheng, 1989).The most recent breakthroughs in implementingactive dynamic gait in bipedal robots have been madeby the Japanese companies Honda (2002) and Sony(2002). Both Honda’s ASIMO (mass 43 kg, height 1.2 m,velocity 0.44 m/s) and Sony’s SDR-3X (mass 5 kg,height 0.5 m, velocity 0.25 m/s) are anthropomorphic