Jumman Hossain

I'm a PhD researcher at UMBC in the Mobile, Pervasive, and Sensor Computing (MPSC) Lab, specializing in reinforcement learning (RL) and machine learning (ML) for autonomous systems and robotics. I focus on optimizing decision-making in complex environments.

During my Summer 2024 internship at Stormfish Scientific Corporation, I developed reinforcement learning models for navigation in XR simulations using AuroraXR®. My work focused on enhancing multi-robot coordination in contested environments, utilizing real-time synchronization between virtual and physical robots through the AuroraXR ROS Bridge.

Email  /  CV  /  Scholar  /  Twitter  /  LinkedIn

Research

My interests lie in reinforcement learning, autonomous navigation, and robotics. I am passionate about advancing autonomous navigation through learning-based optimization techniques. My work focuses on developing algorithms for better understanding and interaction with physical environments—optimizing exploration, navigation, and motion planning, especially in sparse-reward and challenging scenarios.

This paper presents an asymmetric cost-aware navigation method using quasimetric-constrained policy optimization to improve navigation planning in complex, real-world environments with uneven costs and constraints.

This paper introduces a simulation-enhanced approach for realistic navigation planning in multi-agent robotic systems, optimizing strategies for operations in contested and unpredictable environments.

A navigation framework that leverages hierarchical reinforcement learning (HRL) to enable autonomous robots to efficiently explore unknown areas. By building active topological maps and incorporating intrinsic rewards, TopoNav enhances navigation accuracy and adaptability, especially in sparse-reward environments.

This approach enhances maneuver planning by using target-aware visibility estimations and offline reinforcement learning, making it highly suitable for complex and contested environments where maintaining cover is critical.

CoverNav presents a deep reinforcement learning approach for navigation planning that prioritizes cover following in unstructured outdoor environments.

SynchroSim provides an integrated co-simulation middleware that supports coordination and interaction among heterogeneous multi-robot systems. It ensures effective communication and synchronization across diverse platforms, allowing robust development and testing of cooperative robotic applications in dynamic environments.

I utilized source code from here to create this website.