2014 Scholl

Title Investigation on different Approaches for Design of an Active Prosthetic Knee with Elasticity and Damping
Supervisor Mahdy Eslamy
Autor Patrick Scholl (patrick.scholl@stud.tu-darmstadt.de)
Department FB 16 Mechanical Engineering
Last Revision 13. November 2014
Status In Progress
Attention: this site is not finalized! All information are of preliminarily status! Further revisions will come in the near future.

Introduction / Motivation

An average of 67 thousand amputations of the leg or foot in Germany each year can be detected from 2004 to 2012 [1]. A remarkable amount of these amputations is covered by transfemoral amputations, which means that the amputation height is between the knee and the hip. Most of the amputees end up in a wheelchair and thus undergo big changes to their everyday life. In order to ensure an autonomous and active life standard, above-knee prostheses are used.

This thesis deals with different actuation concepts that could be used to ensure low peak power and energy requirement values in active knee prostheses:

1. Direct Drive

2. Serial Elastic Actuator

3. Serial Elastic Damper-Actuator

4. Parallel Elastic Damper-Actuator

By the use of one of the actuation concepts mentionned above human-like motion shall be provided as efficiently as possible. Likewise, by making the leg act as much as possible as a human leg, the user is provided with a natural gait pattern, which reduces harms for the leg without the prosthesis and results in energy-efficient walking. For this issue it is vital to dimension and correlate the parameters of motor, spring and damper.

The aim of this thesis is the design of an active knee prosthesis, which requires least possible restrictions from the amputee. It needs to be light, accurate and efficient to fulfill this expectation.

Approach / Methodology

The aim of this thesis is the design of an active knee prosthesis. In the following chapters the different setups of DD, SEA, SEDA and PEDA should be compared in terms of PP and ER. Furthermore, it needs to be determined what kind of damping is used. The existing ways of damping are completely passive damping, fixed damping values for subsections (as in C-leg) or damping that gets changed all over the gait cycle (as in RHEOKNEE). For each combination of general actuation model and form of damping a Matlab simulation is made in order to calculate energy requirement and peak power values for different kinds of walking. In the simulation the best parameters for the reduction of peak power and energy requirement are determined.

Input data: In earlier measurements of different gaits ([2] and [3]) natural knee angles and torques of level ground and stair walking have been determined. In Matlab simulation the prostheses' torques and angles are assumed to be equal to the knee patterns of healthy subjects. For each setup and each gait type the peak power and energy requirement values for each combination of spring stiffness and damping coefficient are determined. A cost function is used to find the best compromise between energy requirement and peak power and thus the best parameters for each setup at the respective gait type. The formula for the cost function I is displayed below:

I = ((PP_{comb}-PP_{opt})/PP_{opt})² + ((ER_{comb}-ER_{opt})/ER_{opt})²

Simulation Results


For an innovative knee prosthesis the SEA actuation model is not an option due to its high PP and ER in stair descent. In passive damping the PEDA actuation mechanism shows great ER advantages in ER approach of level ground walking compared to the passive damping SEDA. In stair ascent there is no significant difference between any of the passive approaches. For the SEA and passive damping PEDA approach the highest PP occurs in stair descent, while for passive damping SEDA it occurs in stair ascent. As the least possible passive SEDA PP of 2.4606 W/kg is significantly lower than the least possible passive PEDA PP of 3.1756 W/kg, the SEDA geometry is favorable for passive damping. Passive damping SEDA makes clear that although adding a damper to an active prosthesis design results in effort and costs, the advantages of damping are very high. Since active knee prostheses are very expensive and the price of a damping unit would only determine a minor part of its costs, it makes sense to implement it.


Further fields of research:

  • Usage of a damping element with a higher ratio between on- and offstate force
  • Investigation on adding clutches to the actuation concept
  • Inclusion of an electromagnetic energy harvesting damper into PEDA or SEDA concept


  • [1] InEK (2013). Begleitforschung gem. §17 Abs. 8 KHG vom Institut für das Entgeltsystem im Krankenhaus.
  • [2] S. W. Lipfert (2010). Kinematic and dynamic similarities between Walking and running.
  • [3] R. Riener, M. Rabuffetti, and C. Frigo (2002). “Stair ascent and descent at different inclinations”, Gait & posture, vol. 15, no. 1, pp. 32–44.
abschlussarbeiten/bsc/patrickscholl.txt · Zuletzt geändert: 13.09.2016 11:27 von Dario Tokur
GNU Free Documentation License 1.3
Driven by DokuWiki Recent changes RSS feed Valid CSS Valid XHTML 1.0