iamstevepaul.github.io

Follow me on GitHub

Steve Paul

Shows an illustrated sun in light mode and a moon with stars in dark mode.

Research Assistant: Adaptive Design Algorithms, Models, & Systems (ADAMS) Lab

Education

Ph.D. Candidate: Department of Mechanical and Aerospace Engineering (MAE), State University of New York at Buffalo (2020 - Present)

About Me:

I am a Ph.D. candidate in ADAMS Lab, University at Buffalo. My advisor is Dr. Souma Chowdhury. My main research revolves around developing learning-based solutions for complex single/multi-agent combinatorial optimization (planning, scheduling, routing, etc.) with sequential decision-making, considering various environmental and operational uncertainties, in the field of robotics and operations research. Unlike a deterministic problem, where the solutions can be precomputed using expensive solvers, for a dynamic uncertain environment, adaptability is the key to fast efficient decision-making. I make use of the representational power of Graph Neural Networks (GNN) and Topological Data (TDA), for better state information abstraction of various complex single/multi-agent problems (considering centralized and decentralized asynchronous decision-making), to learn optimal policies. I worked as a Software Engineer at Wayfair LLC from June 2018 to December 2019, where I worked on challenging business problems involving Big Data, using various frameworks such as Apache Storm, AKKA, .Net, etc. My passions include reading, working out, hiking, piano, soccer, and astronomy.

Projects

Main Projects

Capacitated Vehicle Routing Problem (CVRP) ->

Tags: Capacitated Vehicle Routing Problem, CVRP, Reinforcement Learning, Graph Neural Networks. Publications: ASME-IDETC 2022

Multi-Robot Task Allocation (MRTA) ->

Tags: Multi-Robot Task Allocation, MRTA, Reinforcement Learning, Graph Neural Networks, Multi-Agent Systems. Publications: IEEE-ICRA 2022 (Nominated for Outstanding Coordination paper)

Multi-Robot Task Allocation - Collective Transport ->

Tags: Multi-Robot Task Allocation, MRTA, Reinforcement Learning, Graph Neural Networks, Multi-Agent Systems. Publications: IEEE-ICRA 2023

Urban Air Mobility (UAM) Fleet Scheduling ->

Tags: Urban Air Mobility, Advanced Air Mobility, Fleet Planning, Reinforcement Learning, Graph Neural Networks, Multi-Agent Systems. Publications: AIAA-AVIATION 2022

Power Network Reconfiguration ->

Tags: Power Networks, Network Reconfiguration, Reinforcement Learning, Graph Neural Networks. Publications: IEEE-TPEC 2022, Nature Communications (Under revision)

Contributed Projects

Urban Air Mobility (UAM) Vertiport Management ->

Tags: Urban Air Mobility, Advanced Air Mobility, Vertiport Management, Reinforcement Learning, Graph Neural Networks, Multi-Agent Systems. Publications: IEEE-IROS 2023, AIAA-SciTech 2023

Concurrent Design for robots for complex urban mission ->

Publications: Under review

Concurrent Design for robots for Multi-Robot Task Allocation ->

Publications: Under review

Email: stevepau@buffalo.edu

LinkedIn

Github

Google Scholar