Learning from Human Video
Failure-Aware Policies
Goal-Conditioned Offline RL
Self-resetting Online RL

Overview

Reinforcement Learning Imitation Learning VLAs / LBMs Deformable Objects Humanoid Locomotion Sim-to-Real
00:00:00
BRISBANE_AU
Robert Lee
Education
PhD Robotics & ML
QUT, Australia (2018-2023)

Experience

Co-Founder & Head of AI Robotics

2026-PRES

Sydekick Robotics

  • Leading AI strategy and development for autonomous robotic systems.
  • Directing end-to-end learning pipelines for general-purpose manipulation.

ML Lead

JUN-DEC 2025

Stealth Humanoid Robotics Co. // Japan

  • Led development of ML pipelines (RL, Imitation Learning, VLAs).
  • Developed RL frameworks for whole-body humanoid locomotion transferring to real hardware.
  • Helped build a strong team of engineers.

Research Engineer

2023-2025

Woven By Toyota // Tokyo, Japan

  • Engineered novel ML approaches for home assistance robots and advanced manipulation.
  • Implemented learning solutions utilizing IL, RL, LLMs/VLMs, and VLAs.
  • Collaborated heavily with Toyota Research Institute on research projects spanning imitation learning, pose estimation, grasping etc.

AI Resident

2021-2022

Everyday Robots & Google X // USA

  • Researched and engineered learning algorithms for everyday, general purpose robots.

Research Intern

2019

Omron Sinic X // Tokyo, Japan

  • Performed research on reinforcement learning for manipulation.
  • Performed software and robotics system engineering (Franka Emika Panda and UR5).

Academic & Research Roles

2016-2019

QUT & CSIRO Data61 // Australia

  • Sessional Academic (Tutor) for Advanced Robotics & Control Systems.
  • Research Assistant applied ML/CV to coastal management datasets.
  • Undergrad Thesis & Intern researching SNN control and human detection.

Research Archives

YEAR TITLE VENUE LINK
2025 IMLE Policy: Fast and Sample Efficient Visuomotor Policy Learning via Implicit Maximum Likelihood Estimation RSS LINK
2025 Learning from 10 Demos: Generalisable and Sample-Efficient Policy Learning with Oriented Affordance Frames CoRL LINK
2025 Can We Detect Failures Without Failure Data? Uncertainty-Aware Runtime Failure Detection for Imitation Learning Policies RSS LINK
2025 ZeroGrasp: Zero-Shot Shape Reconstruction Enabled Robotic Grasping CVPR LINK
2024 Learning fabric manipulation in the real world with human videos ICRA LINK
2024 Gravity-aware grasp generation with implicit grasp mode selection for underactuated hands IROS LINK
2024 DiffusionNOCS: Managing symmetry and uncertainty in sim2real multi-modal category-level pose estimation IROS LINK
2024 Gs-pose: Category-level object pose estimation via geometric and semantic correspondence IROS LINK
2021 Sample-Efficient Learning of Deformable Linear Object Manipulation in the Real World Through Self-Supervision RA-L LINK
2021 TRANS-AM: Transfer Learning by Aggregating Dynamics Models for Soft Robotic Assembly ICRA LINK
2020 Learning Arbitrary-Goal Fabric Folding with One Hour of Real Robot Experience CoRL LINK
2020 A Compact, Cable-driven, Activatable Soft Wrist with Six Degrees of Freedom for Assembly Tasks IROS LINK
2020 Learning Soft Robotic Assembly Strategies from Successful and Failed Demonstrations IROS LINK
2020 Learning Robotic Assembly Tasks with Lower Dimensional Systems by Leveraging Physical Softness and Environmental Constraints ICRA LINK
2020 Contact-based in-hand pose estimation using Bayesian state estimation and particle filtering ICRA LINK
2020 Model-free vision-based shaping of deformable plastic materials IJRR LINK
2019 Evaluating task-agnostic exploration for fixed-batch learning of arbitrary future tasks ACRA LINK
2019 Sim-to-real transfer of robot learning with variable length inputs ACRA LINK