Does it pay to have a damper in a powered ankle prosthesis? A power-energy perspective.

In this paper we investigated on peak power (PP) and energy (ER) requirements for different active ankle actuation concepts that can have both elasticity and damping characteristics. A lower PP or ER requirement is an important issue because it will lead to a smaller motor or battery. In addition to spring, these actuation concepts are assumed to have (passive) damper in series (series elastic-damper actuator SEDA) or parallel (parallel elastic-damper actuator PEDA) to the motor. For SEA (series elastic actuator), SEDA and PEDA, we calculated the required minimum motor PP and ER in different human gaits: normal level walking, ascending and descending the stairs. We found that for level walking and ascending the stairs, the SEA concept, and for descending, the SEDA, were the favorable concepts to reduce required minimum PP and ER in comparison to a DD (direct drive) concept. In SEDA concept, the minimum PP could be reduced to half of what SEA would require. Nevertheless, it was found that spring was always required, however damper showed 'task specific' advantages. As a result, if a simple design perspective is in mind, from PP-ER viewpoint, SEA could be the best compromise to be used for different above-mentioned gaits. For SEDA or PEDA concepts, a controllable damper should be used. In addition, our results show that it is beneficial to select spring stiffness in SEA, based on level walking gait. The PP and ER requirements would increase very slightly for stairs ascending, and to some extent (10.5%) for descending as a consequence of this selection. In contrast, stiffness selection based on stair ascending or descending, increases the PP requirements of level walking more noticeably (17-24%).