Dynamics Movement Primitives-A Framework for Motor Control in Humans and Humanoid Robotics
Research Topic
基本假设：movement sequences consist of sequence segments, executed either in sequence or with partial or completely overlap.
identifying movement primitives (a.k.a. units of actions, basis behaviors, motor schemas, etc.)
这篇文章提出了Dynamic Movement Primitives (DMP).
DMPs are units of action that are formalized as stable nonlinear attractor systems.
key research topic:
- What is a good set of primitives?
- How can they be formalized?
- How can they interact with perceptual input?
- How can they be justed autonomously?
- How can they be combined task specically?
- What is the origin of primitives?
Dynamic Movement Primitives
DMP是受到neurobiology中motor pattern generator (MPG)的启发。
MPG只有moderately modifiable properties，DMP就是添加了一个component that can be adjusted task specifically.
From the view point of movement primitives, kinematic representations are more advantageous than direct motor command coding since this allows for workspace independent planning, and, importantly, for the possibility to superimpose DMP.
文中有一个关键名词"attractor landscape"不知道是什么意思。
In order to accommodate discrete and rhythmic movements, two kinds of DMPs are needed, a point attractive system and a limit system.
The key question of DMPs is how to formalize nonlinear dynamic equations such that they can be flexibly adjusted to represent arbitrarily complex motor behaviors without the need for manual parameter tuning and the danger of instability of the equations.