README.md 1.99 KB
Newer Older
Sylvain Calinon's avatar
Sylvain Calinon committed
1
# Task-parameterized tensor GMM with LQR 
Sylvain Calinon's avatar
Sylvain Calinon committed
2

Sylvain Calinon's avatar
Sylvain Calinon committed
3
### Compatibility
Sylvain Calinon's avatar
Sylvain Calinon committed
4 5 6

	The codes should be compatible with both Matlab and GNU Octave.

Sylvain Calinon's avatar
Sylvain Calinon committed
7
### Usage
Sylvain Calinon's avatar
Sylvain Calinon committed
8

9 10
	Unzip the file and run 'demo_TPGMR_LQR01' (finite horizon LQR), 'demo_TPGMR_LQR02' (infinite horizon LQR) or
	'demo_DSGMR01' (dynamical system with constant gains) in Matlab.  
Sylvain Calinon's avatar
Sylvain Calinon committed
11
	'demo_testLQR01', 'demo_testLQR02' and 'demo_testLQR03' can also be run as additional examples of LQR.
Sylvain Calinon's avatar
Sylvain Calinon committed
12

Sylvain Calinon's avatar
Sylvain Calinon committed
13
### Reference  
Sylvain Calinon's avatar
Sylvain Calinon committed
14

Sylvain Calinon's avatar
Sylvain Calinon committed
15 16
	Calinon, S., Bruno, D. and Caldwell, D.G. (2014). A task-parameterized probabilistic model with minimal intervention 
	control. Proc. of the IEEE Intl Conf. on Robotics and Automation (ICRA).
Sylvain Calinon's avatar
Sylvain Calinon committed
17

Sylvain Calinon's avatar
Sylvain Calinon committed
18
### Description
Sylvain Calinon's avatar
Sylvain Calinon committed
19

Sylvain Calinon's avatar
Sylvain Calinon committed
20 21 22 23 24 25 26
	Demonstration a task-parameterized probabilistic model encoding movements in the form of virtual spring-damper systems
	acting in multiple frames of reference. Each candidate coordinate system observes a set of demonstrations from its own
	perspective, by extracting an attractor path whose variations depend on the relevance of the frame through the task. 
	This information is exploited to generate a new attractor path corresponding to new situations (new positions and
	orientation of the frames), while the predicted covariances are exploited by a linear quadratic regulator (LQR) to 
	estimate the stiffness and damping feedback terms of the spring-damper systems, resulting in a minimal intervention 
	control strategy.
Sylvain Calinon's avatar
Sylvain Calinon committed
27

Sylvain Calinon's avatar
Sylvain Calinon committed
28
### Authors
Sylvain Calinon's avatar
Sylvain Calinon committed
29 30 31 32

	Sylvain Calinon and Danilo Bruno, 2014
	http://programming-by-demonstration.org/
		
Sylvain Calinon's avatar
Sylvain Calinon committed
33 34
	This source code is given for free! In exchange, we would be grateful if you cite the following reference in any 
	academic publication that uses this code or part of it:
Sylvain Calinon's avatar
Sylvain Calinon committed
35 36 37 38 39 40 41 42 43 44 45

	@inproceedings{Calinon14ICRA,
		author="Calinon, S. and Bruno, D. and Caldwell, D. G.",
		title="A task-parameterized probabilistic model with minimal intervention control",
		booktitle="Proc. {IEEE} Intl Conf. on Robotics and Automation ({ICRA})",
		year="2014",
		month="May-June",
		address="Hong Kong, China",
		pages="3339--3344"
	}