ICRA 2022 Workshop on Online Machine Learning-based Control of Lower-Limb Exoskeletons

General Overview

Robotic lower-limb exoskeletons are capable of augmenting human mobility and assisting individuals with mobility impairments. Conventionally, these systems generate parallel joint torques that mimic the user’s underlying biological joint demand during ambulation. Unfortunately, due to the dynamic nature of human movement during daily locomotor activities, it is challenging to develop a control framework that captures the full range of intended movements. However, recent breakthroughs in machine learning (ML) have enabled improved comprehension of the human’s state information in real-time, enabling robust control of these wearable systems during dynamic locomotion. While these ML-based strategies show exciting promise, there remain critical hurdles for these interventions to be deployed to the real world. Challenges include positive feedback loops between actuation and sensing, data size requirements for user-independent models, model robustness to unseen mobility contexts, transitions between locomotion modes, and sensor shifting. In general, there have been few attempts to tackle the critical problem of translating/generalizing laboratory-based ML approaches to real-world, large-scale applications. In this workshop, we will tackle these important challenges from multiple perspectives (both high-level and practical; academic and industrial) and provide roadmaps for future exoskeleton developers to incorporate ML-based controllers for their applications.

Overall Workshop Structure

Our proposed full-day workshop will include 9 in-person sessions (with breaks in between) for a total of 8 hours. All sessions will be live-streamed via video call to enable access for participants joining the conference remotely (except for the poster session). These sessions are seminar talks, panel discussions, short lightning talks, and poster sessions. The organizers believe that an important feature of our workshop is to be inclusive of all backgrounds (demographics, geographical locations, and career status) as different opinions and ideas can contribute to solving our challenging problem in a meaningful way. Therefore, the workshop is structured to be oriented to include both junior and senior researchers and engineers to organically engage and network amongst each other. Additionally, in-between breaks (including lunchtime) will work as networking sessions where researchers can engage with speakers and panelists to extend possible discussions. Overall, the proposed workshop will introduce an energetic and interactive program that ensures diversity while reflecting the state-of-the-art research topic.

Workshop Schedule

Morning Session

  • 08:45 Welcome and workshop overview – Organizers
  • 09:00 Seminar Talk – Aaron Young, Assistant Professor, Georgia Tech
  • 09:30 Seminar Talk – Georgios Evangelopoulos, X (Formerly Google X)
  • 10:00 Coffee Break and Networking Session
  • 10:30 Seminar Talk – Helen Huang, Professor, UNC/NCSU
  • 11:00 Short Lightning Talks
  • 12:00 Lunch and Poster Session

Afternoon Session

  • 14:00 Seminar Talk – Keehong Seo, Principal Engineer, Samsung Electronics
  • 14:30 Seminar Talk – Elliott Rouse, Assistant Professor, University of Michigan
  • 15:00 Panel Discussion – Practical implementation of ML models
  • 15:30 Coffee Break and Interactive Networking Session
  • 16:00 Panel Discussion – Roadblocks and solutions of ML deployment
  • 16:30 Break and Interactive Networking Session
  • 17:00 Concluding Remarks

Organizing Committee

  • Aaron Young, Georgis Institute of Technology
  • Inseung Kang, Georgis Institute of Technology
  • Max Shepherd, Georgia Institute of Technology, Northeastern University
  • Dean Molinaro, Georgia Institute of Technology
  • Georgios Evangelopoulos, X (Formerly Google X)