More Like this

1. Machine Learning Vision and Nonlinear Control Approach for Autonomous Ship Landing of Vertical Flight Aircraft

   Lee, B. ( May 2021 , 77th Annual National Forum of the Vertical Flight Society)

   The paper discusses a machine learning vision and nonlinear control approach for autonomous ship landing of vertical flight aircraft without utilizing GPS signal. The central idea involves automating the Navy helicopter ship landing procedure where the pilot utilizes the ship as the visual reference for long-range tracking, but refers to a standardized visual cue installed on most Navy ships called the ”horizon bar” for the final approach and landing phases. This idea is implemented using a uniquely designed nonlinear controller integrated with machine vision. The vision system utilizes machine learning based object detection for long-range ship tracking, and classical computer vision for object detection and the estimation of aircraft relative position and orientation during the final approach and landing phases. The nonlinear controller operates based on the information estimated by the vision system and has demonstrated robust tracking performance even in the presence of uncertainties. The developed autonomous ship landing system is implemented on a quad-rotor vertical take-off and landing (VTOL) capable unmanned aerial vehicle (UAV) equipped with an onboard camera and was demonstrated on a moving deck, which imitates realistic ship deck motions using a Stewart platform and a visual cue equivalent to the horizon bar. Extensive simulations and flight tests are conducted to demonstrate vertical landing safety, tracking capability, and landing accuracy while the deck is in motion. 

   more » « less
2. A Deep Reinforcement Learning Control Strategy for Vision-based Ship Landing of Vertical Flight Aircraft

   https://doi.org/10.2514/6.2021-3218  

   Lee, Bochan ; Saj, Vishnu ; Benedict, Moble ; Kalathil, Dileep ( July 2021 , AIAA AVIATION 2021 FORUM)

   The paper discusses a deep reinforcement learning (RL) control strategy for fully autonomous vision-based approach and landing of vertical take-off and landing (VTOL) capable unmanned aerial vehicles (UAVs) on ships in the presence of disturbances such as wind gusts. The automation closely follows the Navy helicopter ship landing procedure and therefore, it detects a horizon bar that is installed on most Navy ships as a visual aid for pilots by applying uniquely developed computer vision techniques. The vision system utilizes the detected corners of the horizon bar and its known dimensions to estimate the relative position and heading angle of the aircraft. A deep RL-based controller was coupled with the vision system to ensure a safe and robust approach and landing at the proximity of the ship where the airflow is highly turbulent. The vision and RL-based control system was implemented on a quadrotor UAV and flight tests were conducted where the UAV approached and landed on a sub-scale ship platform undergoing 6 degrees of freedom deck motions in the presence of wind gusts. Simulations and flight tests confirmed the superior disturbance rejection capability of the RL controller when subjected to sudden 5 m/s wind gusts in different directions. Specifically, it was observed during flight tests that the deep RL controller demonstrated a 50% reduction in lateral drift from the flight path and 3 times faster disturbance rejection in comparison to a nonlinear proportional-integral-derivative controller. 

   more » « less
3. AircraftVerse: A Large-Scale Multimodal Dataset of Aerial Vehicle Designs

   https://doi.org/10.5281/zenodo.6525446  

   D., Adam Cobb ; Roy, Anirban ; Elenius, Daniel ; Michael, F. Heim ; Swenson, Brian ; Whittington, Sydney ; D., James Walker ; Bapty, Theodore ; Hite, Joseph ; Ramani, Karthik ; et al ( January 2023 , Zenodo)

   Dataset accompanying code and paper: AircraftVerse: A Large-Scale Multimodal Dataset of Aerial Vehicle Designs

   We present AircraftVerse, a publicly available aerial vehicle design dataset. AircraftVerse contains 27,714 diverse battery powered aircraft designs that have been evaluated using state-of-the-art physics models that characterize performance metrics such as maximum flight distance and hover-time.

   This repository contains:

   • A zip file "AircraftVerse.zip", where each design_X contains:
     • design_tree.json: The design tree describes the design topology, choice of propulsion and energy subsystems. The tree also contains continuous parameters such as wing span, wing chord and arm length.
     • design_seq.json: A preorder traversal of the design tree and store this as design_seq.json.
     • design_low_level.json: The most low level representation of the design. This low level representation includes significant repetition that is avoided in the tree representation through the use of symmetry.
     • Geom.stp: CAD design for the Aircraft in composition STP format (ISO 10303 standard).
     • cadfile.stl: CAD design for the Aircraft in stereolithographic STL file,
     • output.json: Summary containing the UAV's performance metrics such as maximum flight distance, maximum hover time, fight distance at maximum speed, maximum current draw, and mass.
     • trims.npy: Contains the [Distance, Flight Time, Pitch, Control Input, Thrust, Lift, Drag, Current, Power] at each evaluated trim state (velocity).
     • pointCloud.npy: Numpy array containing the corresponding point clouds for each design.
   • corpus_dic: The corpus of components (e.g. batteries, propellers) that make up all aircraft designs. It is structured as a dictionary of dictionaries, with the high level components: ['Servo', 'GPS', 'ESC', 'Wing', 'Sensor', 'Propeller', 'Receiver', 'Motor', 'Battery', 'Autopilot'], containing a list of dictionaries corresponding to the component type. E.g. corpus_dic['Battery']['TurnigyGraphene2200mAh3S75C'] contains the detail of this particular battery.

   Corresponding code for this work is included at https://github.com/SRI-CSL/AircraftVerse. 

   Acknowledgements:

   This material is based upon work supported by the United States Air Force and DARPA under Contract No. FA8750-20-C-0002.  Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the United States Air Force and DARPA.

    

    

   more » « less
4. uavEE: A Modular, Power-Aware Emulation Environment for Rapid Prototyping and Testing of UAV

   M. Theile, O. Dantsker ( January 2018 , Proceedings of IEEE International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA))

   State of the art design and testing of avionics for unmanned aircraft is an iterative process that involves many test flights, interleaved with multiple revisions of the flight management software and hardware. To significantly reduce flight test time and software development costs, we have developed a real-time UAV Emulation Environment (uavEE) using ROS that interfaces with high fidelity simulators to simulate the flight behavior of the aircraft. Our uavEE emulates the avionics hardware by interfacing directly with the embedded hardware used in real flight. The modularity of uavEE allows the integration of countless test scenarios and applications. Furthermore, we present an accurate data driven approach for modeling of propulsion power of fixed-wing UAVs, which is integrated into uavEE. Finally, uavEE and the proposed UAV Power Model have been experimentally validated using a fixed-wing UAV testbed. 

   more » « less
5. Modeling advanced air mobility aircraft in data-driven reduced order realistic urban winds

   https://doi.org/10.1038/s41598-023-50719-8  

   Vuppala, Rohit K. S. S. ; Krawczyk, Zack ; Paul, Ryan ; Kara, Kursat ( January 2024 , Scientific Reports)

   The concept of Advanced Air Mobility involves utilizing cutting-edge transportation platforms to transport passengers and cargo efficiently over short distances in urban and suburban areas. However, using simplified atmospheric models for aircraft simulations can prove insufficient for modeling large disturbances impacting low-altitude flight regimes. Due to the complexities of operating in urban environments, realistic wind modeling is necessary to ensure trajectory planning and control design can maintain high levels of safety. In this study, we simulate the dynamic response of a representative advanced air mobility platform operating in wing-borne flight through an urban wind field generated using Large Eddy Simulations (LES) and a wind field created using reduced-order models based on full-order computational solutions. Our findings show that the longitudinal response of the aircraft was not greatly affected by the fidelity of the LES models or if the spatial variation was considered while evaluating the full-order wind model. This is encouraging as it indicates that the full LES generation of the wind field may not be necessary, which decreases the complexity and time needed in this analysis. Differences are present when comparing the lateral response, owing to the differences in the asymmetric loading of the planform in the full and reduced order models. These differences seen in the lateral responses are expected to increase for planforms with smaller wing loadings, which could pose challenges. Additionally, the response of the aircraft to the mean wind field, the temporal average of the full order model, was misrepresentative in the longitudinal response and greatly under-predicted control surface activity, particularly in the lateral response.

    

   more » « less