MMMCore Documentation

Description

Master Motor Map (MMM) is a conceptual framework for perception, visualization, reproduction, and recognition of human motion in order to decouple motion capture data from further post-processing tasks, such as execution on a real humanoid robot. Employing MMM makes it easy to map motions between different kinematics independently and uniformly as well as to analyze certain dynamic aspects of the considered motion.

Part of this framework is a dynamic model consisting of a particular kinematic structure enriched with pre-defined segment properties (anthropometric data) e.g. mass distribution, segment length, moment of inertia, etc. Moreover, the strategy is to define the maximum number of DoFs that might be used by any visualization, recognition, or reproduction module.

The Master Motor Map Framework is developed by the High Performance Humanoid Technologies (H²T) Lab at the Karlsruhe Institute of Technology.

Design

Human Motion Capture data

All human motion capture modules have an additionally implemented converter module, which transforms the output data to the MMM.

Humanoid Robot Motion Recognition and Reproduction

Modules for visualization, reproduction, and recognition, which need motion data as input, implement an additional converter module, which transforms the data provided in the MMM format to the required input data format.

Illustration of the MMM framework

Framework

We provide a modular C++ framework, which consists of the packages MMMCore and MMMTools.
See the Installing Instructions Page.

MMMCore: The core library

Contains the data structure and kinematic models (e.g. Data I/O , MMM Model, C3D Import), see:

• General Plug-In-Architecture
• MMM Data Format and MMM Sensor Plug-In-Architecture
• MMM Motion Capture Marker Set
• MMM Reference Model and MMM Model Processor
• MMM Motion Tools
• Processing XML files in C++ via RapidXML

MMMTools: The tool library

Contains additional data structure, see:

• MMM Motion Converter
• MMM Motion Segmentation

Contains executable tools for visualization, reproduction and recognition, see:

• MMMViewer as graphical user interface
• MMM Commandline Tools

Acknowledgments

The Master Motor Map (MMM) Framework is used within the European projects KoroiBot (grant agreement no. 611909), WALK-MAN (grant agreement no. 611832), Xperience (grant agreement no. 270273), PACO-PLUS (grant agreement no. 027657), SecondHands (grant agreement no. 643950), TimeStorm (grant agreement no. 641100), and I-SUPPORT (grant agreement no. 643666) funded by the European Union, as well as within the German national projects Collaborative Research Project on Humanoid Robots (SFB 588) and the priority program "Autonomous Learning" (SPP 1527, subprojects REBA and REBA+) funded by the German Research Foundation (DFG).