Research area

I am currently pursuing a Ph.D. program in Aerospace at Cranfield University, focusing on the development of Tactical conflict Resolution solutions based on Machine Learning for Unmanned Aerial Systems Traffic Management systems.
The objective of the research is to explore the capabilities of AI-based solvers for U-Space Service Providers, Suppliers, and Air Traffic Controllers to effectively resolve conflicts between Unmanned Aerial Systems operating within the urban airspace, specifically addressing complex conflict situations and separation assurance with conformance volume constraints.

My Career interests

I am in quest of advancing expertise in research within both Air Traffic Management (ATM) and Unmanned Aircraft System (UAS) Traffic Management (UTM), focusing on the development of innovative services. Additionally, I have a strong interest in data science, avionics, and entrepreneurship, areas which complement and enrich my primary research pursuits

Publications as Main Author

Conferences

Urban Corridor-Based Tactical Conflict Resolution with Flight Plan Adherence and Uncertainty Resilience using a Multi-Agent Reinforcement Learning Solver
- Abstract: This research introduces a Multi-Agent Reinforcement Learning (MARL) system designed for supporting Tactical Conflict Resolution service in shared corridor-based routes for Urban Air Mobility (UAM) operations. The solver incorporates an innovative approach that integrates flight plan adherence into the decision-making process, aligning with the vision of 3D flight corridors as outlined in the latest FAA and NASA Concept of Operations (ConOps). The solver allows for the optimal exploitation of corridor volumes while ensuring safety. The models are trained to be resilient to uncertainties in UAM operations by considering various environmental and sensor-related uncertain factors. The experiment assesses the safety performance of these trained models with, without, or by coupling uncertainties in 0.405 square nautical miles intersecting corridors, as well as evaluating the impact of aircraft performance diversity with up to 12 models and an average of 50 operations per corridor per hour. Under uncertain environment, the solver achieves fewer than one near mid-air collision per flight hour. The outcomes show the importance of comprehensively interpreting the uncertainties from the observation state point of view and demonstrate how to integrate them into an operating environment for effective conflict resolution and separation assurance tasks.
Demonstrating Advanced U-space Services for Urban Air Mobility in a Co-Simulation Environment
- Abstract: The present paper formalises the development of a co-simulation environment aimed at demonstrating a number of advanced U-space services for the Air Mobility Urban- Large Experimental Demonstrations (AMU-LED) project. The environment has a visionary build that addresses Urban Air Mobility (UAM) challenges to support the High/Standard Performance Vehicles (HPV/SPV) operations within a complex urban environment by proposing an integrated solution that packages advanced services from the pre-flight to the in-flight phase in line with ongoing UAM Concept of Operations (ConOps). This setup opts for a holistic approach by promoting intelligent algorithmic design, artificial intelligence, robust serviceability through either virtual and live elements, and strong cooperation between the different services integrated, in addition to sustain interoperability with external U-space Service providers (USSP), Common Information Service providers (CISPs), and Air Traffic Controllers. The prototype has been recently showcased through the AMU-LED Cranfield (UK) demonstration activities.
Application of an autonomous multi-agent system using Proximal Policy Optimisation for tactical deconfliction within the urban airspace
- Abstract: The present paper formalises the development of a Multi-agent Reinforcement Learning (MARL) solver for U- space Service Providers (USSPs) supporting the tactical conflict resolution and exhibited in the Air Mobility Urban - Large Experimental Demonstration (AMU-LED) project. It relies on an Advantage Actor Critic (A2C) model with a Proximal Policy Optimisation (PPO) learning baseline. The application of the autonomous system is demonstrated under a synthetic (with live and virtual) air/unmanned traffic management (ATM/UTM) environment. The Unmanned Aircraft Systems (UASs) are flying in cruise phase at low altitudes, whose respective flight plan generates intersections for enforcing a high collision frequency. The study adopts a step-wise complexity approach of scenarios that confront two agents’ observation methods and showcases a practical case of tactical conflict resolution. The experiments show encouraging deconfliction performance with promising prospects for seeing a such solver deployed.

Find out more research in the bottom links

My current activity

I am enthousiastic to contribute to passionate, ambitious, and innovative projects or roles starting from November 2024.
If you're seeking a dedicated team member, let's connect!

Research area

My Career interests

Publications as Main Author

Conferences

Urban Corridor-Based Tactical Conflict Resolution with Flight Plan Adherence and Uncertainty Resilience using a Multi-Agent Reinforcement Learning Solver

Demonstrating Advanced U-space Services for Urban Air Mobility in a Co-Simulation Environment

Application of an autonomous multi-agent system using Proximal Policy Optimisation for tactical deconfliction within the urban airspace

My current activity