Lucas Manuelli

Technical Lead on the Large Behavior Models team. I was deeply involved in all areas of the project and led work on the data pipelines, ML training and deployment infrastructure, model architecture and training loop along with data collection operations. This work produced the first end-to-end pixels to action neural network policy at Boston Dynamics. Additionally this is one of the first policies in the field to control the full range of humanoid capabilities, including wholebody control and locomotion. For more details see the blog-post I co-wrote https://bostondynamics.com/blog/large-behavior-models-atlas-find-new-footing/

Robotics Research Scientist on Atlas Humanoid Robot team. As part of the Atlas team I am working on perception and manipulation. In particular I am interested in how we can endow robots with manipulation skills similar to those of humans. Although I am have a focus on perception I am interested in all parts of the stack from perception to planning and control. I am driven to develop algorithms that work in the real world on real robot hardware. For a taste of what I've been working on see our latest youtube video (https://youtu.be/-e1_QhJ1EhQ) and behind the scenes video (https://youtu.be/XPVC4IyRTG8).

For more on my research, see https://lucasmanuelli.com Topics: robotics, machine learning, computer vision, 3D perception, self-supervised learning, visuo-motor policy learning.

Implemented an advanced model based grasping algorithm. Deployed the algorithm to a UR10 robot.

I'm working on my PhD in Computer Science at MIT with Russ Tedrake in the Robot Locomotion Group within CSAIL. My research focuses on perception and control for robotic manipulation.

Robot Locomotion Group, CSAIL