I'm an incoming PhD student in the Electrical Engineering and Computer Science department at UC Berkeley. I study controls, intelligent systems, and robotics (CIR) and am advised by Professor Shankar Sastry.

My research interests include modeling innovative drone dynamics, multi-agent games, and optimization. I'm currently working on designing and building an 8 DOF over-actuated drone to compete against a regular drone in aggressive maneuvering at high accelerations. I'm also part of a team designing and testing an RL framework for multi-agent pursuit-evasion games.

Fun fact about me: I served for four years in the US Army as a helicopter mechanic working on the rotor, drive, and engine systems primarily for the CH-47F Chinook and the UH-60L / UH-60M Blackhawks.

Please see my resume: View / Download

[08/2025, UCB] Start PhD!
[06/2025, RSS] Paper accepted to RSS Multi-Robot Systems Workshop at USC
[06/2025, IEEE] Paper under review at the IEEE Global Humanitarian Technology Conference
[02/2025, UCB] Accepted PhD to UC Berkeley!!
[12/2024, UCB] Graduated masters from UC Berkeley!
[12/2024, UCB] Presented capstone project: Neural Machine Translation: Akkadian Cuneiform to English

Evader-Agnostic Team-Based Pursuit Strategies in Partially-Observable Environments
Addison Kalanther, Daniel Bostwick, Chinmay Maheshwari, Shankar Sastry
RSS MRS Workshop 2025

website | code | arxiv

A two-phase neuro-symbolic algorithm for a team of UAV pursuers operating in a partially observable, pursuit-evasion game scenario. By first training adversarial policies offline using deep reinforcement learning and then selecting responses online via a bounded rationality framework, this method improves performance against unpredictable evaders.

Coordinated Autonomous Drones for Human-Centered Fire Evacuation in Partially Observable Urban Environments
Gaby Mendoza, Addison Kalanther, Daniel Bostwick, Emma Stephan, Chinmay Maheshwari, Shankar Sastry
Under Review, IEEE GHTC 2025

website | code | arxiv

A multi-agent coordination framework in which two autonomous UAVs assist human evacuees in real-time during fire evacuations by locating, intercepting, and guiding them to safety under uncertainty. Using a POMDP formulation with panic-informed human behavior and recurrent PPO policies, the UAV team demonstrates significantly improved evacuation efficiency in complex, partially observable environments.

University of California, Berkeley
Electrical Engineering and Computer Science
PhD Controls, Intelligent Systems, and Robotics (CIR)
2025 - Present

Prof: Shankar Sastry

University of California, Berkeley
School of Information
Master of Information and Data Science hello, world!!
2023 - 2024

University of California, Berkeley
School of Letters and Science
Data Science empshasis in Robotics hello, world!!!!!!!
2021 - 2023