Human Aware Interactions and Learning Intern

Toyota Research Institute•Cambridge, MA

3d•$40 - $58•Hybrid

About The Position

Toyota Research Institute (TRI) is seeking a paid research intern to join its Human Aware Interaction and Learning (HAIL) team. The internship is 12 weeks and is a hybrid, in-office role. The research focuses on robot learning and human-robot interaction, contributing to the development of general-purpose robots that can perform a wide variety of dexterous tasks with people. The team studies how large behavior models and other general-purpose machine-learning models can produce robot actions from sensor data and human interactions, using simulation and interactive data to augment real-world robot data. Research may address data-efficient multimodal policies that reason about actions and human presence; hierarchical reasoning over human and shared task states; adapting data-driven policies to human contexts in low-data settings; uncertainty-aware human-robot teaming; use of human-human and synthetic or semi-synthetic data; and shared autonomy with vision-language-action models. The intern will create working code prototypes, interact frequently with team members, run experiments with simulated and physical robots, and participate in publishing work at peer-reviewed venues. The intern should be comfortable working with large static datasets and helping build and maintain a new robot-data corpus.

Requirements

Currently pursuing a Ph.D. or M.S. in Robotics, Computer Science, Mechanical Engineering, or a related field.
Publication record at top-tier robotics or machine-learning conferences, such as RSS, CoRL, ICRA, NeurIPS, ICML, ICLR, CVPR, ICCV, or HRI.
Experience with machine learning and familiarity with large datasets, models, and benchmarks.
Strong software development skills in Python.
A make-it-happen attitude and comfort with fast prototyping and running informative experiments.
Passion for robotics helping people and having real-world, large-scale impact.

Nice To Haves

Hardware experience, especially deploying learned policies on real robotic systems, is strongly preferred.
Experience with C++ is helpful but not required.
Experience conducting user studies involving real robot hardware.
Hands-on experience using machine learning for learned control, including behavior cloning and/or reinforcement learning, for manipulation.
Experience with interactive machine-learning techniques that adapt robot policies based on human feedback or corrections.
Experience with qualitative and quantitative user research, including developing study materials, IRB applications, facilitation and interviewing, managing personally identifiable information (PII), and statistical analysis.
System integration skills, including using state-of-the-art machine-learning tools and databases.

Responsibilities

Develop research in one or more areas at the intersection of robot learning and human-robot interaction.
Create working code prototypes.
Interact frequently with team members.
Run experiments with simulated and physical robots.
Participate in publishing research at peer-reviewed venues.
Work with existing large static datasets and help build and maintain a new corpus of robot data.