> PROGRAM > Workshops
All workshops will be held on Sunday, July 13 (1:00 PM - 6:00 PM) and have limited capacities (maximum 30 participants).
W1: Workshop on machine learning in multibody systems
Lecturers:
Grzegorz Orzechowski (LUT University) and Peter Manzl (University of Innsbruck)
Description: This hands-on workshop provides a comprehensive introduction to machine learning with a focus on deep learning applications for multibody systems. We give an overview and explore the available toolset for practitioners and researchers in the field of multibody dynamics, and practical applications are shown. This workshop is designed for interested researchers with little to no experience in machine learning and neural networks. Sessions will feature both theoretical information and individual coding for small tasks.
- Session 1: Get an introduction to machine learning with an overview of common problems and their associated standard solutions.
- Session 2: Explore supervised learning and how neural networks can be used to learn input-output mapping using labeled data from measurements or simulations.
- Session 3: Apply Reinforcement Learning (RL) to multibody systems and see how the need for labeled data is bypassed in RL by learning from the Interaction with the environment.
- Session 4: See the latest state-of-the-art Large Language Models (LLM), generative text-based models, and their application to the generation of multibody models. In context learning and prompt tuning are applied. The LLMs can be run either locally (if you have a strong laptop) or online using, e.g., ChatGPT or Copilot.
Requirements:
- Participants must bring their own laptop with Anaconda and Python (link) installed. We recommend using Python ≥ 3.10. Please note that some files will be distributed in Jupyter Notebook format.
- The code during the workshop will be distributed through a GitHub repository.
- Ideally, participants should have some basic experience with Python and the additional libraries like exudyn (link), matplotlib (link), jupyter (link), stable-baselines3 (link), and pytorch (link) already installed to get started quickly.
W2: Workshop on spatial operator algebra
Lecturer:
Abhinandan Jain (Karana Dynamics, CA) (formerly JPL, Pasadena, CA)
Description: Minimal coordinate approaches in multibody dynamics offer numerous benefits, including the elimination of redundant coordinates, compatibility with ODE integrators rather than DAE solvers, rapid solution algorithms, and suitability for embedded applications in robotics and control. Despite these advantages, minimal coordinates are under-utilized and often perceived as intricate, challenging, and having limited applicability to general multibody dynamics problems. This workshop will introduce the Spatial Operator Algebra (SOA) methodology for overcoming these challenges associated with minimal coordinate dynamics. SOA employs a concise set of spatial operators to express and reveal the inherent structure of minimal coordinate dynamics models. These operators generate analytical expressions for critical dynamic quantities such as Jacobians, the mass matrix, its inverse, and others. Notably, the operator expressions and analyses exhibit consistency across the broad spectrum of multibody models ranging from simple serial rigid body systems to complex ones with arbitrary size and topology, and with loop constraints as well as flexible bodies.
The workshop will cover key SOA concepts and techniques that can be used to express system level multibody quantities, analytical techniques for exploiting their structure, and methods to develop low-cost scatter and gather recursive computational algorithms. The focus of the workshop will be on covering foundational ideas and theory to facilitate further exploration of more advanced topics and techniques by the attendees.
Requirements:
- The only requirements for the workshop are familiarity with basic kinematics and dynamics concepts for multibody systems.
W3: Workshop on flexible multibody simulation tool Exudyn
Lecturer:
Johannes Gerstmayr (University of Innsbruck)
Description: We are excited to announce a specialized workshop on Exudyn, the easy-to-use tool for simulating flexible multibody systems. Whether you're a seasoned engineer or a researcher in dynamics, this workshop is designed to introduce you to basic architecture of Exudyn as well as hands on modelling of multibody systems with Exudyn. In advanced examples, you can create flexible bodies, including modal reduction, perform parameter optimization, play around with reinforcement learning, or bring your own applications.
- Session 1 - Introduction to Exudyn: Get acquainted with Exudyn's documentation, capabilities and architecture.
- Session 2 - Rigid-Body Tutorial: Work through basic tutorials that demonstrate the setup and simulation of rigid-body systems.
- Session 3 - Advanced Example: Explore a more complex simulation scenario with guidance from experts. Participants are encouraged to bring their own dynamic system problems for discussion and troubleshooting.
Requirements:
- Participants must bring their own laptop with Python and Anaconda installed. We recommend using a recent or the latest versions to ensure compatibility.
- A basic understanding of multibody system dynamics is recommended.
- Ideally, participants should have Exudyn (latest release) installed and have explored the documentation available on the Read the docs: https://exudyn.readthedocs.io
For detailed installation instructions and more about Exudyn, please visit the Exudyn GitHub repository.
