A bird-inspired passive-elastic ankle-toe exoskeleton induces digitigrade (toe) walking in humans
Brekke A. Green Greg S. Sawicki Jonas Rubenson Biomechanics Lab Po�̇ er Lab Biomechanics Lab Penn State University Georgia Tech Penn State University University Park, USA Atlanta, USA University Park, USA [email protected] [email protected] [email protected]
Abstract—A passive-elastic ankle / metatarsophalangeal III. RESULTS & DISCUSSION exoskeleton induced digitigrade (toe) walking in human subjects. A) B) 0.7 0.7 0.7 Engaging the spring reduced the metabolic cost of walking in the Shank pulley Exo with spring engaged 0.6 0.6 0.6 exoskeleton in three of seven participants. Passive contributions to 0.5 0.5 0.5
0.4 0.4 0.4 ankle and toe work likely contribute to energy savings. Load cell 1 0.3 0.3 0.3
0.2 0.2 0.2
Shank 0.1 0.1 0.1 Linear spring
0 Keywords—exoskeleton, walking, bio-inspired, mechanical 0 0 1.1 1.2 1.3 1.4 1.5 1.6 1 1.1 1.2 1.3 1.4 1 1.1 1.2 1.3 1.4 work, passive-elastic toe-down mid-stance toe-off
0.7 0.7
0.6 Rearfoot pulley 0.6 Exo without spring engaged 0.5 0.5 I. INTRODUCTION 0.5
0.4 Ankle joint 0.4 Load cell 2 0.4 0.3 0.3 Passive-elastic lower-limb exoskeletons aim to reduce the 0.3
0.2 0.2 Toe joint/pulley 0.2 metabolic cost of walking by taking advantage of elastic springs 0.1 0.1 Rearfoot 0.1
Cable termination 0 0 0 0.9 1 1.1 1.2 1.3 1 1.1 1.2 1.3 1.4 1.1 1.2 1.3 1.4 1.5 1.6 1.7 that provide both assistive passive torque and mechanical work. heel-strike mid-stance toe-off [1]. We propose that further advances in passive-elastic assistive exoskeletons can be achieved by drawing inspiration from non- C) Forefoot ANKLE AnkleANKLE MPTOE TOE human species that exemplify elastic gait mechanics. Unlike 60 60
20 ) 20 dorsiflex. dorsiflex. humans, the majority of cursorial species walk and run on their 40 40 0 deg 0 ( toes (digitigrade). This mode of locomotion facilitates elastic 20 20 −20 −20 plantar flex. Angle plantar flex. energy storage and return in muscle-tendon units that cross both− 40 −40 0 0 0 20 0 40 20 60 40 80 60 100 80 100 0 20 0 40 20 60 40 80 60 100 80 100 the ankle and the metatarsophalangeal (MP) joint. For example,100 100 dorsiflex. dorsiflex. 0 0 the ostrich has been shown to produce fifty percent of the 0 (Nm) 0 −20 −20 positive mechanical work of the stance phase of running through−100 −100 plantar flex. plantar flex. −200 −40 elastic recoil at the tarso-metatarsophalangeal joint [2]. Moment −200 −40 0 20 0 40 20 60 40 80 60 100 80 100 0 20 0 40 20 60 40 80 60 100 80 100
The purpose of the present study was to design and test a 50 50 200 200 gen. gen.
(W) 0 0 biologically-inspired human exoskeleton prototype that 0 0 −50 −50 −200 emulates the storage and return of elastic energy at both the −200 −100 −100 Power abs. abs. −400 −400 −150 −150 ankle and MP joints typical of the gait mechanics of efficient 0 20 0 40 20 60 40 80 60 100 80 100 0 20 0 40 20 60 40 80 60 100 80 100 % Stride % Stride non-human cursorial species. Fig. 1. A) EXO design; B) Center of pressure and ground reaction forces during the stride; C) Gait mechanics with EXO spring assist (solid blue), II. METHODS without spring assist (dashed blue) and the EXO spring contribution (red). The exoskeleton (EXO) consisted of three segments: (1) a shank frame; (2) a rear-foot frame mounted to a running shoe; The EXO resulted in a center of pressure anterior to the MP and (3) a fore-foot plate that was mounted to the running shoe joint. The EXO spring contributed to 20% and 15% of the peak ankle an MP moments, respectively, and 25% of the ankle work. under the toe box (Fig. 1). A spring cord ran across a pulley Metabolic cost was reduced by engaging the EXO spring in situated on the rear-foot frame behind the ankle and continued three subjects, remained constant in one and increased in three across a second pulley on the rear-foot frame to terminate on subjects. A two-joint passive ankle-toe EXO can repurpose the the fore-foot plate. Two parallel linear steel springs (7 kN/m MP joint to function more spring-like during walking. each) were positioned in the spring cord line behind the calf. Two strain gauge load cells (Omega) measured spring force. REFERENCES [1] S.H.. Collins, M.B Wiggin, G.S. Sawicki “Reducing the energy cost of Metabolic cost was measured in seven subjects (Cosmed K4 human walking using an unpowerede exoskeleton”. Nature, vol 522, / K5) during treadmill walking (1.25 m s-1) without the EXO; pp.212–5, 2015. with the EXO but with no elastic component engaged; and with [2] J. Rubenson, D.G. Lloyd, D.B. Heliams, T.F. Besier, P.A. Fournier. the EXO fully functional. Overground joint mechanics were “Adaptations for economical bipedal running: the effect of limb structure recorded in the same conditions at a similar speed (1.25 m s-1 ± on three-dimensional joint mechanics.” J. R. Soc. Int, vol 8, pp. 740–55, 2011. 5%; 8-camera Motion Analysis Corp; two AMTI force plates).