Aidan Curtis

I'm a fourth year PhD student at MIT EECS advised by Leslie Kaelbling, Tomás Lozano-Pérez, and Joshua Tenenbaum. I am grateful to be supported by the NSF Graduate Research Fellowship and the Harold E. Edgerton Memorial Fellowship. My recent internship experiences include IBM, MERL, and Boston Dynamics.

I'm interested in discovering the cognitive mechanisms in humans that lead to flexible embodied intelligence and applying those to real-world robots. My current research centers around learning abstractions and handling uncertainty for long-horizon robotic manipulation and navigation planning.

Email  /  Google Scholar  /  Twitter  /  Github

profile photo
Publications
Bayes3D: fast learning and inference in structured generative models of 3D objects and scenes
Nishad Gothoskar*, Matin Ghavami*, Eric Li* Aidan Curtis, Michael Noseworthy, Karen Chung, Brian Patton William T. Freeman, Joshua B. Tenenbaum, Mirko Klukas, Vikash K. Mansinghka
Arxiv Preprint
Paper / Code

We present Bayes3D, an uncertainty-aware perception system for structured 3D scenes, that reports accurate posterior uncertainty over 3D object shape, pose, and scene composition in the presence of clutter and occlusion. Bayes3D delivers these capabilities via a novel hierarchical Bayesian model for 3D scenes and a GPU-accelerated coarse-to-fine sequential Monte Carlo algorithm.

Task-Directed Exploration in Continuous POMDPs for Robotic Manipulation of Articulated Objects
Aidan Curtis, Leslie Kaelbling, Siddarth Jain
ICRA, 2023
Paper

In this paper, we propose STRUG, an online POMDP solver capable of handling domains that require long-horizon planning with significant task-relevant and task-irrelevant uncertainty. We demonstrate our solution on several temporally extended versions of toy POMDP problems as well as robotic manipulation of articulated objects using neural perception.

Visibility-Aware Navigation Among Movable Obstacles
Aidan Curtis*, Jose Muguira-Iturralde*, Yilun Du, Leslie Pack Kaelbling, Tomás Lozano-Pérez
ICRA, 2023
Paper / Code

In this paper, we examine the problem of visibility aware robot navigation among movable obstacles (VANAMO). A variant of the well-known NAMO robotic planning problem, VANAMO puts additional visibility constraints on robot motion and object movability.

PG3: Policy-Guided Planning for Generalized Policy Generation
Ryan Yang*, Tom Silver*, Aidan Curtis, Tomas Lozano-Perez, Leslie Pack Kaelbling
IJCAI, 2022
Paper / Code

In this work, we study generalized policy search-based methods with a focus on the score function used to guide the search over policies. The main idea is that a candidate policy should be used to guide planning on training problems as a mechanism for evaluating that candidate.

Let's Handle It: Generalizable Manipulation of Articulated Objects
Aidan Curtis*, Zhutian Yang*
ICLR Workshop, 2022
Paper / Code

This workshop paper describes our award-winning solution to the Sapien Maniskill Manipulation Challenge. We present a framework for building generalizable manipulation controller policies that map from raw input point clouds and segmentation masks to joint velocities.

Long-Horizon Manipulation of Unknown Objects via Task and Motion Planning with Estimated Affordances
Aidan Curtis*, Xiaolin Fang*, Leslie Pack Kaelbling, Tomás Lozano-Pérez, Caelan Reed Garrett
ICRA, 2022
Paper / Video / Code

We present a strategy for designing and building very general robot manipulation systems involving the integration of a general-purpose task-and-motion planner with engineered and learned perception modules that estimate properties and affordances of unknown objects.

Discovering State and Action Abstractions for Generalized Task and Motion Planning
Aidan Curtis, Tom Silver, Joshua B Tenenbaum, Tomas Lozano-Perez, Leslie Pack Kaelbling
AAAI, 2022
Paper / Code

Generalized planning accelerates classical planning by finding an algorithm-like policy that solves multiple instances of a task. Here we apply generalized planning to hybrid discrete-continuous task and motion planning.

Map Induction: Compositional spatial submap learning for efficient exploration in novel environments
Sugandha Sharma, Aidan Curtis, Marta Kryven, Josh Tenenbaum, Ila Fiete
ICLR, 2022
Paper / Code

Humans are expert explorers. In this work, we try to understand the computational cognitive mechanisms that support this efficiency can advance the study of the human mind and enable more efficient exploration algorithms.

Planning with learned object importance in large problem instances using graph neural networks
Tom Silver*, Rohan Chitnis*, Aidan Curtis, Joshua Tenenbaum, Tomas Lozano-Perez, Leslie Pack Kaelbling
AAAI, 2021
Video / Code / Paper

Real-world planning problems often involve hundreds or even thousands of objects, straining the limits of modern planners. In this work, we address this challenge by learning to predict a small set of objects that, taken together, would be sufficient for finding a plan.

A Spatiotemporal Map of Reading Aloud
Oscar Woolnough, Cristian Donos, Aidan Curtis, Patrick S Rollo, Zachary J Roccaforte, Stanislas Dehaene, Simon Fischer-Baum, Nitin Tandon
JNeurosci, 2021
Paper

Reading words aloud is a foundational aspect of the acquisition of literacy. We used direct intracranial recordings in a large cohort to create a holistic yet fine-grained map of word processing, enabling us to derive the spatiotemporal neural codes of multiple word attributes critical to reading

Flexible and efficient long-range planning through curious exploration
Aidan Curtis, Minjian Xin, Dilip Arumugam, Kevin Feigelis, Daniel Yamins
ICML, 2020
Paper / Code / Website

A core problem of long-range planning is finding an efficient way to search through the tree of possible action sequences. Here, we propose the Curious Sample Planner (CSP), which fuses elements of TAMP and DRL by combining a curiosity-guided sampling strategy with imitation learning to accelerate planning.

Threedworld: A platform for interactive multi-modal physical simulation
Chuang Gan, Jeremy Schwartz, Seth Alter, Martin Schrimpf, James Traer, Julian De Freitas, Jonas Kubilius, Abhishek Bhandwaldar, Nick Haber, Megumi Sano, Kuno Kim, Elias Wang, Damian Mrowca, Michael Lingelbach, Aidan Curtis, Kevin Feigelis, Daniel M Bear, Dan Gutfreund, David Cox, James J DiCarlo, Josh McDermott, Joshua B Tenenbaum, Daniel LK Yamins
NeurIPS, 2022
Paper / Website / Code

We introduce ThreeDWorld (TDW), a platform for interactive multi-modal physical simulation.

HealthSense: Software-defined mobile-based clinical trials
Aidan Curtis, Amruta Pai, Jian Cao, Nidal Moukaddam, Ashutosh Sabharwal
Mobicom, 2019
Paper

We take a software-inspired viewpoint of clinical trial designs to enable expressibility of complex trials, composability with diverse devices and services while maximally maintaining simplicity for a clinical research user.

Saccadic Corruption of Long Range Cerebral Connectivity Metrics
Aidan Curtis, Kiefer Forseth, Oscar Woolnough, Cihan Kadipasaoglu, Nitin Tandon
Biorxiv, 2022
Paper

Top-down visual object recognition processes driven by the human orbitofrontal cortex (OFC) have been proposed to facilitate rapid processing of images in higher-level visual regions. We found that trials lacking the saccade artifact on scalp EEG also lacked low gamma band PLV increase in iEEG. This work illustrates the importance of eliminating confounding saccadic artifacts.

Other Projects
Wildfire Prevention and Management using Deep Reinforcement Learning
Aidan Curtis*, William Shen*
Paper / Project / Code

We use Deep Reinforcement Learning to train AI agents which are able to combat wildfires. This page demonstrates the videos of our learned policies. Please see our paper for more details.

Short Term Spatiotemporal Video Prediction on Sports via Convolutional LSTMs
Aidan Curtis*, Victor Gonzalez*
Paper

Predicting short term video dynamics has many useful applications in self-driving cars, weather nowcasting, and model-based reinforcement learning. In this project we provide an in-depth analysis of the available models for video prediction and their strengths and weaknesses in predicting natural sequences of images.

Actor Critic Reinforcement Learning in 2D and 3D
Aidan Curtis, Kevin Feigelis
Blog

The project assesses the efficacy of finetuning off different neural architectures trained on ImageNet for Actor-Critic reinforcement learning in 2D and 3D environments. We find a combination of semantic and spatial information results in the best few-shot performance.

Wireless Recorder for Intracranial Epileptic Seizure Monitoring
Aidan Curtis, Sophia D’Amico, Andres Gomez, Benjamin Klimko, Zhiyang Zhang
Paper / Website / Video / Code

In this project we design and build a wireless intracranial neural recording system that uses sparse coding compression to efficiently transmit neural data.


website source code