Tractable Adaptability

Ethan K. Gordon

Postdoc, University of Pennsylvania

PhD 2023, University of Washington

Active Learning for Contact-Rich Assistive Manipulation

Ethan at Lehigh:

Physically Assistive Robots (PARs)

“If I can have a robot do it, I can learn to adapt to it, but it would be me feeding me, and that would be huge”

 

Tyler Schrenk

1985-2023

The Promise of PARs:

  • Empowerment
  • Independence

Tractable Adaptability

How can robots efficiently learn, during deployment, how to manipulate previously-unseen objects?

What is needed for pHRI?

Contact-Rich Manipulation

  • Sliding to clean the spoon and bowl
  • Shaking to smoothen
  • In-Mouth Hand-Off
    (vision-denied)

 

Online Adaptation

  • Bite Size Adjustment

What is needed for pHRI?

Online Adaptation

  • Totally Different Food
  • Multi-bite: different shapes for each bite

 

There is no time for

re-training!

The Technology/Application Cycle

Support

Inform

Physically Assistive Robots (PARs)

Active Learning in Contact

The Technology/Application Cycle

Support

Inform

Physically Assistive Robots (PARs)

Active Learning in Contact

HRI 2025
HRI 2020
HRI 2024
NIPS 2018
RA-L 2022
ICRA 2023
ICRA 2022
Preprint 2026
ISRR 2019
IROS 2020
ICRA 2023
CoRL 2023

Autonomous Bite Acquisition

Problem: Bite Acquisition

Online Bite Acquisition Challenges

  • Large Action Space (whole trajectory) OR Sparse Reward
  • Unknown Dynamics / State Transition: food simulation is hard!

Solution: Data-Driven Policy-Space Simplification

(Gordon, CoRL 2023)

Learn This Online!

Emergent Behavior

Wiggling

Tilting

High Pressure

Scooping

Multimodal Online Learning

  • High-Dimension State Space: Foods are really diverse!

Solution: Haptic Policy Regularization

(Bhattacharjee, R-AL 2019); (Gordon, IROS 2020); (Gordon, ICRA 2021)

Optimize Jointly

Active Learning in Contact

Dynamic Object System Identification

Choose:

  • Robot Trajectory \(r[t]\)

Measure:

  • Contact Boolean \(c_t\)

  • Contact Normal \(\hat{n}_t\)

  • Proprioception

Find:

  • Object Geometry \(\theta^*\)

  • Object Pose \(x^*_T\)

Learning with a Violation-Implicit Loss

Information Maximization In Action

Expected Information Gain (EIG)

Learn; Compute

Observed Info \(\mathcal{I}\)

Sample + Simulate

Expected Fisher Info \(\mathcal{F}\)

\(\max EIG := \log\det\left(\mathcal{F}\mathcal{I}^{-1} + \mathbf{I}\right)\)

Choose actions where simulated, expected Fisher info is distinct from Observed info.

Robot-Assisted Feeding

User Studies Capture Diversity

(Bhattacharjee, HRI 2020)

User Studies Capture Metrics 

(Bhattacharjee, HRI 2020)

Trade-off between autonomy (with chance of error) and high-effort manual control.

What errors are tolerable?

Community-Based Participatory Design

(Gordon, HRI Companion 2024)

Community-Based Participatory Design

(Nanavati, HRI 2025)

Research Plans

Safe Active Exploration in Contact
(ActSafe, ICLR 2025)

Research Plans

Dressing "Acquisition"
(TOORAD, Autonomous Robots 2019)

Research Plans

Multi-Function PARs
?

Thank you!

DAIR Lab

Amal Nanavati

Tractable Adaptability

Ethan K. Gordon

Postdoc, University of Pennsylvania

PhD 2023, University of Washington

Active Learning for Contact-Rich Assistive Manipulation

Structure Through Expert-Defined Heuristics

  • Qualitative Taxonomy of Single-Utensil Bite Acquisition
  • Convert to Action Schema:
    • \(\mathbb{R}^{14} \times SO(3) \times S^2\)
    • Force and Torque Thresholds

Benefits: Interpretable, Continuity (Similar Numbers \(\rightarrow\) Similar Action)

(Gordon, CoRL 2023); (Bhattacharjee, R-AL 2019)

Data-Driven Discretization

(Gordon, CoRL 2023)

Exploration vs. Exploitation: Contextual Bandits

(Gordon, IROS 2020); SPANet from (Feng, IJRR 2019)

Incorporating Haptic Information

(Gordon, ICRA 2021); (Bhattacharjee, R-AL 2019)

Classification with 50ms of 6DOF F/T Data

\(l_t = c_t^T\theta^* + \epsilon_\theta = p_t^T\phi^* + \epsilon_\phi\)

Optimize both simultaneously, regularizing them against each other.

Exploration vs. Exploitation: Contextual Bandits

(Gordon, ICRA 2021)

Active Learning for pHRI: Spinning the Flywheel

User-Informed:

Metrics

Priorities

Limitations

Contact-Rich Active Learning:

Model-Based

Policy Simplification

Multimodal Sensing

Community-Based:

System Design

Implementation

Pain Point Identification