Aaron Ray

PhD Student | MIT EECS

firstlast at mit.edu

I am a PhD student in the MIT Spark Lab, working on the intersection of spatial perception, language grounding and planning for Robotics. I am very fortunate to be advised by Prof. Luca Carlone.

My current research focuses on unifying perception and planning representations for mobile robots. I work on using 3D scene graphs as a symbolic abstraction of sensor data for real-time, on-robot perception, and grounding natural language commands to structured planning representations using these 3D scene graph data structures.

Previously, I completed my masters researching multi-robot coordination algorithms for a variety of information-gathering tasks in the Distributed Robotics Lab with Prof. Daniela Rus. I applied a combination of high level planning and nonlinear model predictive control techniques to increase the capabilities of teams of drones.

Before MIT, I was an undergraduate student at Brown University where I built race cars, drones, and found the love of my life.

Publications

Scene Graphs

I worked on a variety of projects related to 3D scene graphs and symbolic planning. 3D scene graphs provide a discrete abstraction of continuous high dimensional sensor data. Symbolic planning allows for scalable, verifiable planning and robot execution.

Task and Motion Planning in Hierarchical 3D Scene Graphs

Aaron Ray*, Christopher Bradley*, Luca Carlone, Nicholas Roy. International Symposium of Robotics Research (ISRR), 2024. Watch Video
Language-Grounded Hierarchical Planning and Execution with Multi-Robot 3D Scene Graphs

Jared Strader, Aaron Ray, Jacob Arkin, Mason B Peterson, Yun Chang, Nathan Hughes, Christopher Bradley, Yi Xuan Jia, Carlos Nieto-Granda, Rajat Talak, Chuchu Fan, Luca Carlone, Jonathan P How, Nicholas Roy. International Symposium on Experimental Robotics (ISER), 2025. Watch Video
Hydra-multi: Collaborative online construction of 3D scene graphs with multi-robot teams

Yun Chang, Nathan Hughes, Aaron Ray, Luca Carlone. 2023 International Conference on Intelligent Robots and Systems (IROS). Watch Video
Structured 3D Scene Queries with Graph Databases

Aaron Ray, Luca Carlone. Learned Robot Representations workshop @ RSS 2025.

Videography Drones

In the Distributed Robotics lab, I focused on integrating discrete planning with nonlinear model predicative control (MPC).

Free-space ellipsoid graphs for multi-agent target monitoring

Aaron Ray, Alyssa Pierson, Daniela Rus. International Conference on Robotics and Automation (ICRA) 2022.
Multi-robot task assignment for aerial tracking with viewpoint constraints

Aaron Ray, Alyssa Pierson, Hai Zhu, Javier Alonso-Mora, Daniela Rus. International Conference on Intelligent Robots and Systems (IROS) 2021.
Robust flight navigation out of distribution with liquid neural networks

Makram Chahine*, Ramin Hasani*, Patrick Kao*, Aaron Ray*, Ryan Shubert, Mathias Lechner, Alexander Amini, Daniela Rus. Science Robotics, 2023. Watch Video
Closed-form continuous-time neural networks

Ramin Hasani, Mathias Lechner, Alexander Amini, Lucas Liebenwein, Aaron Ray, Max Tschaikowski, Gerald Teschl, Daniela Rus. Nature Machine Intelligence, 2022.

Drone Hardware

I have also worked on a variety of drone hardware projects, including the PiDrone, a low-cost aerial educational platform used in a college-level robotics course at Brown University. In this course, students build, program, and test drones to create autonomous aircraft capable of visual localization and position maintenance using onboard sensors. Additionally, as part of SPARK Lab, I developed a soft aerial manipulator with a perception pipeline for robust object grasping, featuring a novel tendon-actuated soft gripper for fast and adaptive grasping.

Pidrone: An autonomous educational drone using raspberry pi and python

Isaiah Brand, Josh Roy, Aaron Ray, John Oberlin, Stefanie Tellex. International Conference on Intelligent Robots and Systems (IROS) 2018.
High-speed aerial grasping using a soft drone with onboard perception

Samuel Ubellacker, Aaron Ray, James M Bern, Jared Strader, Luca Carlone. npj Robotics 2024. Watch Video

Open Source Software Project Contributions

Omniplanner

Infrastructure for composing multiple types of planning paradigms. Omniplanner provides an interface to solving DSG-grounded planning problems with a variety of solvers and grounding mechanisms.

Spot Tools

Spot robot tools (navigation, picking up objects, opening doors).

Awesome-DCIST-T4

Curated list of papers and resources focused on the MIT DCIST stack intended to keep pace with the anticipated surge of research in the coming months.