NSF: NRI: INT: Safe Wind-Aware Navigation for Collaborative Autonomous Aircraft in Low Altitude Airspace

Last updated on May 28, 2022

Award Information NSF Award Number 1925147

Investigators

Abstract This project will promote the progress of science, and advance the national health, security and prosperity by conducting fundamental research needed to enhance safety and efficiency of aircraft operations in low altitude urban environments. Safe and seamless inclusion of unmanned aircraft such as delivery drones, and manned aircraft such as air taxis into urban airspace requires building traffic management systems that account for the performance of aircraft, pilots, and autopilots under a variety of atmospheric conditions. Aircraft operating in low altitude urban environments are subject to turbulent wind gusts that cannot be accurately predicted using current techniques. This research project will address this challenge by enabling the progress of science across multiple disciplines, including meteorology, human-robot interaction, machine learning, data-driven modeling, robotics, and instrumentation. Desired research outcomes will significantly impact many important applications, including micro-weather prediction, turbulent plumes of pollutants and emissions, drone package delivery, and air taxis. The inter-disciplinary nature of the project will better prepare next-generation students and engineers and the synergistic activities will broaden the participation of underrepresented groups in research.

The project aims to validate the hypothesis that ‘in-time’ gust awareness by a pilot or an autopilot, can enhance safety, efficiency and robustness of future autonomous aircraft operations in low altitude airspace. Towards this objective, the research team will investigate novel learning tools to model piloting behaviors, design safe and efficient wind aware path planning algorithms, and importantly, construct short-term gust forecast models with wind measurements. The team will develop a high-fidelity simulation framework that integrates turbulence modeling, guidance, navigation, control, and pilot-aircraft interface to demonstrate autonomous and remotely-piloted aircraft flying through urban canopies with improved predictability and increased endurance. A recommendation system that facilitates pilot-aircraft interactions will be produced and demonstrated through both simulations and experiments.

This award reflects NSF’s statutory mission and has been deemed worthy of support through evaluation using the Foundation’s intellectual merit and broader impacts review criteria.

Large-Eddy Simulation LES Unmanned Aircraft Systems UAS Gust Turbulence Urban Environment Machine Learning ROM LSTM NSF Award Number 1925147
Kursat Kara
Kursat Kara
Assistant Professor, Mechanical and Aerospace Engineering

Dr. Kara is the principal investigator of the Kara Aerodynamics Research Laboratory at Oklahoma State University. He teaches the Fundamentals of Aerodynamics, Unsteady Aerodynamics, Computational Fluid Dynamics, and Quantum Computing. Previously, he was an assistant professor at Khalifa University, where he received the Faculty Excellence Award for Outstanding Teaching in 2015.

Publications

A non-intrusive reduced order model using deep learning for realistic wind data generation for small unmanned aerial systems in urban spaces
Realistic wind data are essential in developing, testing, and ensuring the safety of small unmanned aerial systems in operation. We present a non-intrusive reduced order modeling (NIROM) approach to replicate realistic wind data and predict wind fields.
Rohit Vuppala, Kursat Kara
PDF Cite Code Project DOI
A non-intrusive reduced order model using deep learning for realistic wind data generation for small unmanned aerial systems in urban spaces
Investigation of Airflow around Buildings using Large Eddy Simulations for Unmanned Air Systems Applications
We use Large-Eddy Simulation to understand the unsteady and highly coherent turbulent flow structures produced by buildings in neutral atmospheric boundary layer flow. Furthermore, we demonstrate a non-intrusive machine learning methodology to predict flow fields to augment safe wind-aware navigation systems for Unmanned Aerial Vehicles as a first step toward safely integrating UAS into existing aerial infrastructure.
Tyler Landua, Rohit Vuppala, Kursat Kara
PDF Cite Project Video DOI
Investigation of Airflow around Buildings using Large Eddy Simulations for Unmanned Air Systems Applications
Realistic Wind Data Generation for Small Unmanned Air Systems in Urban Environment using Convolutional Autoencoders
We attempt to create accurate wind data for an urban environment using high-fidelity CFD data from Large Eddy Simulations (LES) and Convolutional Auto-Encoders (CAE) for non-linear surrogate modeling. The non-linear surrogate model extracts underlying non-linear modes from the high-resolution data snapshots, and the LSTM network trains on these specific modes.
Rohit Vuppala, Kursat Kara
PDF Cite Project
Realistic Wind Data Generation for Small Unmanned Air Systems in Urban Environment using Convolutional Autoencoders
A Novel Approach in Realistic Wind Data Generation for The Safe Operation of Small Unmanned Aerial Systems in Urban Environment
A single building setup in neutral atmospheric conditions is considered a test case to demonstrate the method. The method relies on using Large Eddy Simulation data from a computational fluid dynamics simulation and a non-intrusive Reduced Order Modeling approach (ROM) coupled with Recurrent Neural Networks like Long Short Term Memory (LSTM).
Rohit Vuppala, Kursat Kara
PDF Cite Project Video DOI
A Novel Approach in Realistic Wind Data Generation for The Safe Operation of Small Unmanned Aerial Systems in Urban Environment