Tomás Lozano-Perez

School of Engineering Professor in Teaching Excellence, Department of Electrical Engineering and Computer Science

Computer Science and Artificial Intelligence Laboratory

AI Robotics

Computer Vision

Machine Learning

External links

Google Scholar

Missions

Embodied Intelligence

Synopsis

Tomás Lozano-Perez is an MIT professor interested in integrating task, motion and decision-theoretic planning for robotic manipulation.

Biography

Tomás Lozano-Perez is the School of Engineering Professor in Teaching Excellence at MIT, and a principal investigator at the Computer Science and Artificial Intelligence Laboratory. His research has been in robotics (configuration-space approach to motion planning), computer vision (interpretation-tree approach to object recognition), machine learning (multiple-instance learning), medical imaging (computer-assisted surgery) and computational chemistry (drug activity prediction and protein structure determination from NMR and X-ray data). His current research is aimed at integrating task, motion and decision-theoretic planning for robotic manipulation. Lozano-Perez has received an IEEE Robotics Pioneer Award and a Presidential Young Investigator Award.

Tomás Lozano-Perez

Synopsis

Biography

Publications

Keypoint Abstraction using Large Models for Object-Relative Imitation Learning

Combining Planning and Diffusion for Mobility with Unknown Dynamics

Trust the PRoC3S: Solving Long-Horizon Robotics Problems with LLMs and Constraint Satisfaction

Keypoint Abstraction using Large Models for Object-Relative Imitation Learning

Combining Planning and Diffusion for Mobility with Unknown Dynamics

Trust the PRoC3S: Solving Long-Horizon Robotics Problems with LLMs and Constraint Satisfaction

Embodied Uncertainty-Aware Object Segmentation

Practice Makes Perfect: Planning to Learn Skill Parameter Policies

DiMSam: Diffusion Models as Samplers for Task and Motion Planning under Partial Observability

Compositional Diffusion-Based Continuous Constraint Solvers

Embodied Lifelong Learning for Task and Motion Planning

Monte Carlo Tree Search in Continuous Spaces Using Voronoi Optimistic Optimization with Regret Bounds

Online Replanning in Belief Space for Partially Observable Task and Motion Problems

PDDLStream: Integrating Symbolic Planners and Blackbox Samplers

Learning to Guide Task and Motion Planning Using Score-space Representation

In the News

Helping robots practice skills independently to adapt to unfamiliar environments

New technique helps robots pack objects into a tight space

Solve Every Problem With Extreme Metadata

Helping robots practice skills independently to adapt to unfamiliar environments

New technique helps robots pack objects into a tight space

Solve Every Problem With Extreme Metadata

AI helps household robots cut planning time in half

Automating the search for entirely new “curiosity” algorithms

Four from MIT named 2017 Association for Computing Machinery Fellows.

NASA's humanoid robot lands at CSAIL