Difference between revisions of "VVV09 Kinematic, my friend"

From Wiki for iCub and Friends
Jump to: navigation, search
Line 2: Line 2:
  
 
== FAQ like ==
 
== FAQ like ==
 
+
'''Why learn the model, we already have an analytic method ?'''
Why learn the model, we already have an analytic method ?  
 
 
* A human like approach.
 
* A human like approach.
 
* To validate the feasibility of such kind of learning.
 
* To validate the feasibility of such kind of learning.
  
Why the forward model ?
+
'''Why the forward model ?'''
 
* To keep all the information of the model (i.e infinity of pseudo-inverse of the Jacobian, null space computation).
 
* To keep all the information of the model (i.e infinity of pseudo-inverse of the Jacobian, null space computation).
  
Why the kinematic ?
+
'''Why the kinematic ?'''
 
* Because the geometric model is not easily invertible.
 
* Because the geometric model is not easily invertible.
  
What kind of algorithm is using ?
+
'''What kind of algorithm is using ?'''
 
[[Image:pc_learning.jpg|right|192px|thumbnail|[[The learning attitude]]!]]
 
[[Image:pc_learning.jpg|right|192px|thumbnail|[[The learning attitude]]!]]
 
* LWPR, (Locally Weighted Projection Regression) [Vijayakumar 2005]. A powerfull online learning algorithm based on multiple linear regression computation in the space the most adapted (PLS).
 
* LWPR, (Locally Weighted Projection Regression) [Vijayakumar 2005]. A powerfull online learning algorithm based on multiple linear regression computation in the space the most adapted (PLS).
  
What are the results ?
+
'''What are the results ?'''
 
* The model is learned and uses "successfully" on the robot.
 
* The model is learned and uses "successfully" on the robot.
  
What about the "keep all the information" ?
+
'''What about the "keep all the information" ?'''
 
* With the learned model we are able to control two end-effectors with a priority on one of them.
 
* With the learned model we are able to control two end-effectors with a priority on one of them.
  
So, what did you do here ?
+
'''So, what did you do here ?'''
 
* I tried to learn the model ... without the model but with the vision.
 
* I tried to learn the model ... without the model but with the vision.
  
And ?
+
'''And ?'''
 
* I have learned the model by attended motor babbling (As we play with the "end effector" of a baby).
 
* I have learned the model by attended motor babbling (As we play with the "end effector" of a baby).
 
* I "wanted" to be able to track two objects (my end effector and another object ... But it is apparently not so easy :p) in order to complete the learning of the model still as if we were playing with a child.
 
* I "wanted" to be able to track two objects (my end effector and another object ... But it is apparently not so easy :p) in order to complete the learning of the model still as if we were playing with a child.

Revision as of 10:27, 29 July 2009

The aim subject is the learning of the forward kinematic model.

FAQ like

Why learn the model, we already have an analytic method ?

  • A human like approach.
  • To validate the feasibility of such kind of learning.

Why the forward model ?

  • To keep all the information of the model (i.e infinity of pseudo-inverse of the Jacobian, null space computation).

Why the kinematic ?

  • Because the geometric model is not easily invertible.

What kind of algorithm is using ?

  • LWPR, (Locally Weighted Projection Regression) [Vijayakumar 2005]. A powerfull online learning algorithm based on multiple linear regression computation in the space the most adapted (PLS).

What are the results ?

  • The model is learned and uses "successfully" on the robot.

What about the "keep all the information" ?

  • With the learned model we are able to control two end-effectors with a priority on one of them.

So, what did you do here ?

  • I tried to learn the model ... without the model but with the vision.

And ?

  • I have learned the model by attended motor babbling (As we play with the "end effector" of a baby).
  • I "wanted" to be able to track two objects (my end effector and another object ... But it is apparently not so easy :p) in order to complete the learning of the model still as if we were playing with a child.