Positive feedback in powered exoskeletons: improved metabolic efficiency at the cost of reduced stability?
Proc. of the ASME International Design Engineering Technical Conference & Computers and Information in Engineering Conference, Las Vegas, NV, Sept. 4-7, 2007.

James A. Norris, Anthony P. Marsh, Kevin P. Granata, Shane D. Ross


A broad objective of many lower extremity exoskeletons is to allow the wearer to expend less of their own energy for locomotion. Existing exoskeleton control algorithms are based on positive feedback. Forces are generated to augment movement initiated by the wearer. Positive feedback, however, can have destabilizing effects in dynamic systems. In fact, stability in these lower extremity exoskeletons is achieved by relying on the wearer's neuromuscular system. Relying on the wearer to maintain stability may increase metabolic demand, which is counter productive to increasing efficiency. Thus, the goal of this study was to measure how a simple form of positive feedback that augments ankle push-off power affects both metabolic efficiency and dynamic walking stability. We developed a pair of powered ankle-foot orthoses (PAFOs) similar in design to Ferris, et al. (J. Appl. Biomech. 21, 189-197, 2005). Nine young healthy adults (23.3 +/- 1.6 years) walked on a treadmill in the PAFOs under two conditions: (1) with and (2) without push-off power assistance. Metabolic energy expenditure was calculated using indirect calorimetry. Walking stability was quantified using techniques for studying stability of dynamic system trajectories. The maximum Lyapunov exponent for assessing local dynamic stability, and the maximum Floquet multiplier magnitude for assessing orbital stability were calculated from foot and shank kinematics for each condition. Greater Lyapunov exponents and Floquet multipliers indicate decreased stability. Walking with mechanically generated push-off power increased metabolic efficiency (2.58 +/- 0.39 to 2.97 +/- 0.38, p<0.01), did not affect local dynamic stability (0.14 +/- 0.02 to 0.14 +/- 0.02, p=0.77), but decreased orbital dynamic stability (0.43 +/- 0.03 to 0.48 +/- 0.06, p=0.05). This study provides evidence that positive feedback can negatively affect stability. Further investigations into understanding stability of movement will be necessary for the design of controllers for powered lower extremity exoskeletons.

Back to Papers | Home