Institut für Sportwissenschaft

Diese Seite ist nicht editierbar. Sie können den Quelltext sehen, jedoch nicht verändern. Kontaktieren Sie den Administrator, wenn Sie glauben, dass hier ein Fehler vorliegt.
====== 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

  * Power and Energy Requirements in Active Prosthetic Feet in Different Actuation Mechanisms
  * Control of Active Foot Prostheses
      
4. Conclusions 


===== Motivation =====


**Lower extremity amputations done in Germany:** 
    * More than 60000 per year (Institut für das Entgeltsystem im Krankenhaus, 2012)

**Important Question:**
    * How can engineering help the amputees to regain their normal locomotion performance?



===== 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]


[{{ :biomechanik:abschlussarbeiten:ch1fpassifoot.jpg?400 |Some passive foot prostheses from Ossur and Ottobock}}]

[{{ :biomechanik:abschlussarbeiten:ch1fankpowampsoun.png?400 |Ankle joint power: Amputated vs. Non-amputated Side (Postema1997)}}]


The above figure shows that there is "Lack of Power Generation" on the amputated side

[{{ :biomechanik:abschlussarbeiten:ch1factivefeet.jpg?400 |Active Prosthetic feet}}]





===== Performance of Active Prosthetic feet =====  

•	Reduce metabolic cost of transfer (CoT)  (Au2009)

•	Emulate ankle power similar to the human ankle 



[{{ :biomechanik:abschlussarbeiten:ch5fpowersacti.jpg?400 |The desired and real ankle joint power for PAKO experiments}}]


===== Challenges for Active Prosthetic feet =====  
 

  * Power and  Energy Requirements
         * As low as possible  

  * Control
         * How should the robot know what to do for each gait? 


**Why low Power and Energy Requirements?**

      * Lower Power Requirement -->  Smaller Motor, Less Weight

      * Lower Energy Requirement -->  Smaller Battery, Less Chargings  

**One Solution is to focus on:**
       * Biologically-Inspired Actuation Mechanisms 
===== Mechanical model of the biological actuator (Muscle) =====  



[{{ :biomechanik:abschlussarbeiten:musclehill.png?400 |Mechanical model of the biological actuator}}]


[{{ :biomechanik:abschlussarbeiten:muscle-models.png?400 |Different Muscle Models based on Hill's works}}]


**Important Question:**

  * What is the importance of those Components and their Configurations in Power and Energy requirements of active foot prostheses?

  * First Actuation Scheme could be as simple as DD (Direct Drive) or SEA (Series Elastic Actuator)

[{{ :biomechanik:abschlussarbeiten:ddandsea.png?400 |DD actuator vs SEA actuator}}]


===== Peak Power and Energy Requirement =====  

 


[{{ :biomechanik:abschlussarbeiten:peakpowerprogram.png?400 | Procedure to obtain Peak Power and Energy Requirement}}]


  * **Peak Power and Energy Requirement: DD vs. SEA**

[{{ :biomechanik:abschlussarbeiten:ppanderwalkingsea.png?400 |Peak Power and Energy Requirement DD vs SEA walking 1.55 m/s}}]


  * **Comparison of the Peak Power Requirement**      
  


[{{ :biomechanik:abschlussarbeiten:comparisonppddvssea.png?400 |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:**

  * Why PP and Energy Requirements of SEA are less than DD?


  * **Comparison of Motor Force and velocity (DD vs SEA)** 

[{{ :biomechanik:abschlussarbeiten:motvelseadd.png?400 |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**


[{{ :biomechanik:abschlussarbeiten:ppallspeeds.png?400 |PP changes w.r.t. spring stiffness in SEA actuator}}] 


  * As it is seen in this figure, there is only one stiffness that really minimizes the PP requirement in SEA actuators (this holds for all speeds)

===== Parallel Elastic Element (PEE) =====  

  * In this section we consider the effect of adding a parallel elastic element on the PP and Energy requirements of active foot prostheses. 


[{{ :biomechanik:abschlussarbeiten:pee1.png?200 |Parallel Elastic Element (PEE) could be also used in a powered ankle}}]


**Different Approaches for Parallel Elastic Element**

  * We could have different types of actuators when we include a parallel element 

[{{ :biomechanik:abschlussarbeiten:peedifferenttypes.png?400 |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**


[{{ :biomechanik:abschlussarbeiten:peecomparisonppande.png?400 |effect of PEE on PP and Energy Requirement}}]


  * In comparison to SEA, the PE (parallel element) reduced the PP requirement but increased the Energy requirement (figure for Walking)

**Motor Velocity, force and power pattern: SEA vs. SEA+UPS**

[{{ :biomechanik:abschlussarbeiten:motvelforpee.png?400 |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) =====  

  * In this section we consider the effect of adding damping element on PP and Energy requirement in active foot prostheses.

[{{ :biomechanik:abschlussarbeiten:dampingelemnt.png?230 |Damping Element (DE) to be used in an active foot prosthesis}}]


**Parallel or Series Damping Element (DE)**

  * Damping element could be in series or parallel with the motor: 

[{{ :biomechanik:abschlussarbeiten:pedaandseda.png?400 |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**

  * In addition we also consider the level walking and stair ascending and descending and investigate the power and energy requirements for these gaits 


[{{ :biomechanik:abschlussarbeiten:dampingdiffegaits.png?400 |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**

[{{ :biomechanik:abschlussarbeiten:dampingresults.png?400 |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


===== Bi-articular Actuation in Active Foot Prosthesis =====   


**Main ankle plantar-flexors**

There are two main ankle plantar-flexors: Soleus and Gastrocnemius, Soleus is mono-articular and Gastrocnemius is bi-articular 



[{{ :biomechanik:abschlussarbeiten:solandgas.png?400 |Main ankle plantar-flexors: Soleus and Gastrocnemius}}]


**How to Benefit from Bi-articular Actuation in Active Foot Prosthesis**

  * How to select appropriate stiffness: The Weighted Sum of PP and Energy Requirement
[{{ :biomechanik:abschlussarbeiten:weightedsumappr.png?400 |The Weighted Sum of PP and Energy Requirement, lambda is a weighting factor}}]

**Comparison of PP and Energy requirements: SEA vs. SEA+G**

[{{ :biomechanik:abschlussarbeiten:weightedsumresults.png?400 |Comparison of PP and Energy requirements: SEA vs. SEA+G}}]


  * As it is seen the weighted sum approach, defines the PP and Energy requirement of an active foot prosthesis in a way that both PP and Energy requirements could be determined by the designer so that they are not far away from minim values. 


===== Control Challenges in Active Foot Prostheses =====  




[{{ :biomechanik:abschlussarbeiten:controlchallenges.png?400 |Slave controller is for DC motor control, Master controller is for Gait Identification}}]

**Lab experiments with PAKO** 
  * Results for walking 0.5 m/s (PAKO: Powered Ankle Knee Ortho-prosthesis)


[{{ :biomechanik:abschlussarbeiten:pakoresults.png?400 |Lab experiments with PAKO, motor force, velocity and ankle power and angle }}]