2015 Eslamy
Title | Emulation of Ankle Function for Different Gaits through Active Foot Prostheses: Actuation Concepts, Control and Experiments |
Leitung | Prof. Dr. phil. Andre Seyfarth |
Autor | Mahdy Eslamy (m.eslamy at gmail dot com) |
Letzte Bearbeitung | Mar. 8th 2015 (Under Review) |
indexmenu_n_1
Outline
1. Motivation: Why Research in this Field?
2. Background and Previous Outcomes
3. Contributions of this Thesis
4. Conclusions
Motivation
Lower extremity amputations done in Germany:
Important Question:
Previous Works
Performance of Passive Prosthetic Feet
slower self-selected speeds than able-bodied people [Waters1976]
expend more metabolic energy [Molen1973]
kinematic difference between amputated side and sound side [Sanderson1997]
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/ch1fpassifoot.jpg?w=400&tok=f9a1c3)
Some passive foot prostheses from Ossur and Ottobock
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/ch1fankpowampsoun.png?w=400&tok=a590bb)
Ankle joint power: Amputated vs. Non-amputated Side (Postema1997)
The above figure shows that there is „Lack of Power Generation“ on the amputated side
• Reduce metabolic cost of transfer (CoT) (Au2009)
• Emulate ankle power similar to the human ankle
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/ch5fpowersacti.jpg?w=400&tok=dfad0b)
The desired and real ankle joint power for PAKO experiments
Challenges for Active Prosthetic feet
Why low Power and Energy Requirements?
One Solution is to focus on:
Mechanical model of the biological actuator (Muscle)
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/musclehill.png?w=400&tok=4ad354)
Mechanical model of the biological actuator
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/muscle-models.png?w=400&tok=383014)
Different Muscle Models based on Hill's works
Important Question:
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/ddandsea.png?w=400&tok=28aa88)
DD actuator vs SEA actuator
Peak Power and Energy Requirement
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/peakpowerprogram.png?w=400&tok=041aa7)
Procedure to obtain Peak Power and Energy Requirement
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/ppanderwalkingsea.png?w=400&tok=7ca758)
Peak Power and Energy Requirement DD vs SEA walking 1.55 m/s
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/comparisonppddvssea.png?w=400&tok=910d94)
Comparison of the Peak Power Requirement
As it is seen, the PP (peak power) requirement of SEA actuator is reduced about 58% in comparison to DD (see also the pictures of DD and SEA).
Important Question:
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/motvelseadd.png?w=400&tok=6f41cd)
Comparison of Motor velocity (DD vs SEA)
As it is seen, the motor velocity in SEA is less than DD in the push-off time (the red zone), as power is the multiplication of force and velocity, consequently the PP requirements will also be less.
In addition, as a considerable amount of power is provided by spring, the motor power requirement in general is less than DD and because of this matter the Energy requirement is also less than DD.
Importance of Stiffness for PP Requirement
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/ppallspeeds.png?w=400&tok=18cba4)
PP changes w.r.t. spring stiffness in SEA actuator
Parallel Elastic Element (PEE)
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/pee1.png?w=200&tok=1d47c3)
Parallel Elastic Element (PEE) could be also used in a powered ankle
Different Approaches for Parallel Elastic Element
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/peedifferenttypes.png?w=400&tok=11e653)
Different Approaches for Parallel Elastic Element PS: Parallel Spring UPS: Unidirectional Parallel Spring
Peak Power and Energy Requirement in case of Adding a Parallel Element
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/peecomparisonppande.png?w=400&tok=a38e4a)
effect of PEE on PP and Energy Requirement
Motor Velocity, force and power pattern: SEA vs. SEA+UPS
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/motvelforpee.png?w=400&tok=09991c)
Motor Velocity, force and power pattern: SEA vs. SEA+UPS
The PP of SEA+UPS reduced because the peak force of SEA+UPS is less than SEA, however as in SEA+UPS the motor is doing work against the parallel spring in swing phase, this requires power and consequently energy requirement increases in comparison to SEA.
Conclusion on Parallel Elastic Element
To reduce disadvantages for Energy Requ. in PS/UPS approach, we should control when and how long the UPS is engaged
No case was found in which PS had a better result than UPS (in terms of PP and Energy Requ.)
SEA could be an acceptable compromise for actuation
Damping Element (DE)
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/dampingelemnt.png?w=230&tok=0dee39)
Damping Element (DE) to be used in an active foot prosthesis
Parallel or Series Damping Element (DE)
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/pedaandseda.png?w=400&tok=f96a89)
Damping element could be in series or parallel with the motor, PEDA: Parallel Elastic Damping Actuator, SEDA: Series Elastic Damping Actuator
Damping Element (DE) in Active Foot Prostheses: Different Gaits
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/dampingdiffegaits.png?w=400&tok=6ab16e)
we also consider the level walking and stair ascending and descending to investigate the power and energy requirements for these gaits
Peak Power and Energy Requirements
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/dampingresults.png?w=400&tok=97030f)
Peak Power and Energy Requirements: PEDA vs SEDA vs SEA
Conclusions on Damping Element:
Stiffness is required for all gaits
Damping shows on-off behavior (as seen in the picture for level walking and stair ascent the damping should be off however for stair descent it is required)
SEA could be an acceptable compromise for actuation
Main ankle plantar-flexors
There are two main ankle plantar-flexors: Soleus and Gastrocnemius, Soleus is mono-articular and Gastrocnemius is bi-articular
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/solandgas.png?w=400&tok=ba9a70)
Main ankle plantar-flexors: Soleus and Gastrocnemius
How to Benefit from Bi-articular Actuation in Active Foot Prosthesis
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/weightedsumappr.png?w=400&tok=520c74)
The Weighted Sum of PP and Energy Requirement, lambda is a weighting factor
Comparison of PP and Energy requirements: SEA vs. SEA+G
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/weightedsumresults.png?w=400&tok=227e03)
Comparison of PP and Energy requirements: SEA vs. SEA+G
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/controlchallenges.png?w=400&tok=a95aec)
Slave controller is for DC motor control, Master controller is for Gait Identification
Lab experiments with PAKO
![](http://wiki.ifs-tud.de/_media/biomechanik/abschlussarbeiten/pakoresults.png?w=400&tok=77514a)
Lab experiments with PAKO, motor force, velocity and ankle power and angle