04_movement_primitives.md

4. Movement Primitives

Currently, we provide three Movement Primitives, including Dynamic Movement Primitives(DMPs), Probablistic Movement Primitives(ProMPs) and Probablistic Dynamic Movement Primitives(ProDMPs). We also provide two Movement Primitives Interfaces(Dynamic based and Probablistic based), which can be used to define the custom Movement Primitives. You can register your custom MPs to the MPFactory and create the MP instance given configuration.

The main features of the MP Factory and MP Interfaces are as follows:

Classes	Main Functions	Description
MPFactory	MPFactory.init_mp	Create an MP instance given configuration.
MPInterface		Abstract Basic Class for Deterministic Movement Primitives
	MPInterface.update_inputs	Update MPs parameters
	MPInterface.get_trajs	Get movement primitives trajectories given flag
	MPInterface.learn_mp_params_from_trajs	Abstractmethod for learning parameters from trajectories
ProbabilisticMPInterface		Abstract Basic Class for Probablistic Movement Primitives
	MPInterface.update_inputs	Update MPs parameters
	MPInterface.get_trajs	Get movement primitives trajectories given flag, including trajectories mean and distribution
	MPInterface.sample_trajectories	Sample trajectories from MPs
	MPInterface.learn_mp_params_from_trajs	Abstractmethod for learning parameters from trajectories

 

4.1 Dynamic Movement Primitives

We provide a DMPs demo to show how to create a DMPs instance and visualize the corresponding result.

To run the demo, you can run the following code:

from mp_pytorch import demo
demo.test_dmp()

 

4.2 Probablistic Movement Primitives

We provide a ProMPs demo to show how to create a ProMPs instance and visualize the corresponding result.

To run the demo, you can run the following code:

from mp_pytorch import demo
demo.test_promp()
demo.test_zero_padding_promp()

 

4.3 Probablistic Dynamic Movement Primitives

Probablistic Dynamic Movement Primitives(ProDMPs) is a recently presented Method, which combing the Dynamic and Probablistic properties of Movement Primitives from a unified perspective.

We provide a ProDMPs demo to show how to create a ProDMPs instance and visualize the corresponding result. To run the demo, you can run the following code:

from mp_pytorch import demo
demo.test_prodmp()

Back to Overview