Model Integrated Computing MIC

1
Model Integrated Computing (MIC)
TEAM-1
Olawale Fapohunda Daanish Khan Sam Odviwri K.
Omair Muhi
2
Overview

• Introduction Background
• MIC in industry
• Sample application
• Future directions
• Conclusion

3
Some Definitions

• Models Formal structure representing selected
  aspects of the engineering artifact and its
  environment
• Modeling Usage of abstractions (models) as
  programming elements
• Model-Based Systems Systems employing concepts
  of modeling
• Provides basis for solving complex systems

4
Motivation

• Software interaction with systems
• Embedded systems e.g. car brakes
• Tight integration
• Software and physical systems must evolve
  together

5
Model Integrated Computing

• Captures software architecture and environment
• Dependencies and constraints among various
  modeling views
• Rapid design and implementation of systems

6
Model Integrated Computing (contd)

• Extending the scope and use of models
• Modeling paradigms matched to the needs of the
  domain engineers
• Provides framework for developing domain
  artifacts for computer-based systems

7
Domain Specific Modeling

• 1

8
MIC Design Flow

• Models composed of reusable components
• Can be built and tested independently
• Doesnt require knowledge of models outside
  focus area

9
MIC How its used

• Industrial Applications
• Industrial Robotics
• Vehicle Design
• Non-Programming related fields
• Technicians dont need to know about C/C
• Wide variety of concerns
• Performance
• Safety, etc

10
MIC The specifics

• Develop Domain Specific Modeling Environment
  (DSME)
• Your reusable models for your application
• Examples NI Labview, Matlab/Simulink
• Embed constraints into the model (OCL)
• Specific model behaviors
• Some buzzwords
• Multi-View Modeling of Cross-Cutting Constraints
• Meta-Code Generation
• Framework Coupling and Merging

11
GME Generic Modeling Environment

• Toolkit for creating domain-specific design
  environments (DSDE DSME)
• Specify meta-model using UML/OCL
• Meta-model interpreted into configuration for
  GME, becomes DSME
• Popular GMEs include
• GME2000
• Dome by Honeywell Research

12
MIC in Industrial Robotics

• Hard real-time constraints
• MIC reduces debugging time for
• Non-linear dynamics
• Kinematics models
• Sam will cover in more detail

13
An Application of MIC in Robot Control systems

• Area of Application Platform independent
  automated code synthesis of dynamics models
• Benefits
• Flexibility in model specification and usage
• Dynamic adaptation and reconfiguration of
• model.
• Much less debug iterations
• Less chance of introducing bugs when
• making changes

14
How is this achieved?

• Mechanical parts modelling
• Multi Body System (MBS)
• a finite set of primitive components
• joints, springs, links and drives.
• Dedicated algorithm
• compute torque, plan path, calibrate, etc.
• Modularity design
• Parts and components
• model Database

15
Putting all together...

• Model component interaction via ports
• Transmission objects
• A manipulator object serves as a center control.
• Creates and connects transmission objects
• serves as the domain specific code generator
• performs model re-configuration

16

• Left A generic 3 freedom robot
• Right A possible MBS model representation

17
How does it run?

• Execution
• a set of routines declared in the transmission
• interface (domain specific).
• Transmission objects
• contain concrete implementation
• Run based on interpreted or generated code by
  manipulator.

18
Stages involved in the domain specific code
generation
19
Future directions of MIC

• Scalability of supporting tools
• distributed simultaneous access to models
• version control
• challenge respecting interdependencies

19
20
Future directions of MIC (contd)

• Improving model-interpreter generation
• writing model interpreters viewed as a bottleneck
• exploit GGT to solve bottleneck problem

21
Conclusion

• MIC concept
• GME and industry applications
• Sample application robot manipulator
• Future directions

22
References

• R. Hopler, P. J. Mosterman, Model Integrated
  Computing in Robot Control to Synthesize
  Real-time Embedded Code," Institute of Robotics
  and Mechatronics
• J. Sztipanovits, G. Karsai, Model-Integrated
  Computing, Integrated Engineering, April 1997
• A. Ledeczi, G. Balogh, Z. Molnar, P. Volgyesi, M.
  Maroti, Model Integrated Computing in the
  Large, Institute for Software Integrated Systems
  (ISIS), 2003
• A. Agrawal, Graph Rewriting And Transformation
  (GReAT) A Solution For The Model Integrated
  Computing (MIC) Bottleneck , ISIS, Vanderbilt
  University
• Hernandez.F, Bangalore.P, Gray.J, Reily.K , A
  Graphical Modeling Environment for the Generation
  of Workflows for the Globus Toolkit , Workshop
  on Component Models and Systems for Grid
  Applications , June-July 2004.
• A. Ledeczi et al. Metamodel Composition in the
  Generic Modeling Environment, Institute for
  Software Integrated Systems (ISIS), Vanderbilt
  University, 2003

23

• QUESTIONS?