Search NASA⌕ Search

Engineering topics

Stefan Schuet

Publications and source records attributed to Stefan Schuet.

14 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS2021

An Approach to Reasoning Service Migration in Data and Reasoning Fabric (DRF) Implementation

In this paper we consider service migration problem for Data and Reasoning Fabric (DRF) enabled airspace operations assuming a fixed cloud/edge infrastructure with allocated computing, storage and power resources, where cloud/edge servers, and communication stations are in a wired connected network, while vehicles use a wireless network for communication. The objective is to automatically select the best location for the requested service execution, which achieves minimum cost while satisfying the user quality of service (QoS) and available resources constraints. To this end, estimates of the response time, consumed energy and total cost are defined for each potential compute location. A mixed-integer linear program is then formulated and solved to identify optimal compute locations given QoS constraints, network infrastructure limitations, with worst-case vehicle positioning. The approach is applied to trajectory re-planning use case to avoid a collision with an emergency vehicle in real time.

Air mobility↗
NASA NTRS2020

A Modeling Approach for Handling Qualities and Controls Safety Analysis of Electric Air Taxi Vehicles

The combination of modern advances in electric propulsion, fly-by-wire controls, autonomy, and increasing demand for short range air taxi operations, is currently producing an outburst of vehicle designs more diverse than ever before. Advanced software tools are needed to support the rapid and safe introduction of any design into the airspace, including the safety of the deployed flight control system and vehicle handling qualities. This paper presents a methodology for building air taxi vehicle models with distributed electric propulsion for use in analyzing flight control system safety at the conceptual design level.The approach builds on existing software tools capable of outputting aeromechanics-based linear perturbation models for Vertical Take-off and Landing vehicles with multiple rotors. Rotor torque inputs are then converted into equivalent voltage control inputs, and the linear state and input dynamics matrices are modified to include electric motor dynamics with common parameters for direct-current electric motors. The linear perturbation dynamics are then stitched across multiple operating points into a quasi-Linear Parameter Varying model that covers the full flight envelope. A Model Predictive Controller is developed for use with the full envelope model, and a tradeoff analysis between handling quality and motor requirements is demonstrated using a six passenger NASA air taxi reference design.

Urban Air Mobility↗
NASA NTRSDate not supplied

An Approach to Reasoning Service Migration in Data and Reasoning Fabric (DRF) Implementation

In this paper we consider service placement problem for Data and Reasoning Fabric (DRF) enabled airspace operations assuming a fixed cloud/edge infrastructure with allocated computing, storage and power resources, where cloud/edge servers, and communication stations are in a wired connected network, while vehicles use a wireless network for communication. The objective is to automatically select the best location for the requested service execution, which achieves minimum cost while satisfying the user quality of service (QoS) and available resources constraints. To this end, we estimate for each potential location the response time, consumed energy and total cost; formulate an optimization problem for cost minimization given the users QoS constraints and network infrastructure limitations; and solve it using nonlinear programming tools. The approach is applied to trajectory re-planning use case to handle a no-fly zone contingency in real time.

Vahram Stepanyan↗
NASA NTRSDate not supplied

A Gaussian Process Enhancement to Linear Parameter Varying Models

Simulation and analysis for modern engineering systems now routinely requires the merging of multiple disciplines, physical-domains, time-scales, and data sets — all at ever increasing levels. These capabilities are especially needed in the domain of Advanced Air Mobility, where rapidly emerging vehicle designs are significantly more complex, while having to be both cost-effective and safe. To meet these engineering challenges, machine learning methods are an attractive option for merging models and data across multiple areas while providing uncertainty quantification and maintaining computational efficiency. This paper examines the use of Gaussian process machine learning to generalize and enhance the commonly used class of quasi-Linear Parameter Varying models for fast full-envelope simulation while also supporting control system design and analysis with model uncertainty. Gaussian process machine learning is selected because it: can fuse multiple data sets, enables an easy trade-off between data fitting and smoothing, provides model uncertainty quantification, scales well with increasing complexity, and does not generally require starting from a large training data set. To demonstrate the benefits of the approach, a robust stability analysis with Gaussian process uncertainty is shown for a NASA reference design of an electric quad-rotor air-taxi concept vehicle with motor parameter uncertainty.

Gaussian Process↗
NASA NTRSDate not supplied

Evaluation of Heave Disturbance Rejection and Control Response Criteria on the Handling Qualities Evaluation of Urban Air Mobility (UAM) eVTOL Quadrotors Using the Vertical Motion Simulator

The first piloted handling qualities study of an urban air mobility (UAM) vehicle leveraging the Vertical Motion Simulator (VMS) at NASA Ames Research Center was conducted in Spring 2021. The VMS provides a unique capability to reduce risk by assessing and iterating control designs. Minimal sources currently exist to provide performance and handling qualities data for large, rotor speed-controlled vehicles outside of the software environment. The study compares multiple handling qualities performance configurations for rotor speed and blade pitch-controlled variants of a six-passenger quadrotor conceptual design model developed by the NASA Revolutionary Vertical Lift Technology (RVLT) Project. Additionally, both ADS-33 and a tailored set of performance standards (notionally representing the agility required of a UAM mission) are examined under conditions with and without light turbulence. Preliminary results did show significant variation in ratings based on the set of standards utilized, controller tuning to either Level 1 or boundary Level 1/ Level 2 conditions, and presence or lack of turbulence. A custom approach and landing maneuver was also designed to bring these evaluation tasks together in a more comprehensive application.

Handling Qualities↗
NASA NTRSDate not supplied

Analysis of Handling Qualities and Power Consumption for Urban Air Mobility (UAM) eVTOL Quadrotors with Degraded Heave Disturbance Rejection and Control Response

A piloted handling qualities study of urban air mobility (UAM) electric vertical take-off and landing (eVTOL) quadrotors was performed utilizing the Vertical Motion Simulator (VMS) facility at NASA Ames Research Center. Rotor speed and variable pitch-controlled variants of a six-passenger conceptual design vehicle were assessed with different levels of degradation to control response and disturbance rejection bandwidth (DRB) in the heave axis. In previous work, preliminary trends across several handling quality rating categories reflected the effects of these degradations. Additionally, the impact of using different test standards and turbulence on the ratings were discussed. This paper elaborates on those results, but also provides insight into unexpected trends observed during the study including: a disharmony in attitude response, subpar ratings for the baseline Level 1 performance vehicle, and excessive drift and yaw couplings observed in a lateral reposition maneuver. Moreover, shortcomings of the handling quality scales and comparisons of power consumption among the vehicles in the various test conditions are presented.

Handling Qualities↗
NASA NTRSDate not supplied

Integrated Handling Qualities Safety Analysis For Conceptual Design of Urban Air Mobility Vehicles

The recent emergence of distributed electric propulsion Vertical Takeoff and Landing (VTOL) aircraft has created a rapid introduction of new design concepts with unique stability and control characteristics. A challenge to realizing the full potential of these vehicles for urban transportation is to gain public acceptance which is largely driven by flight safety. This paper describes ongoing research to address the feasibility of integrating flying qualities safety metrics into conceptual design of VTOL Urban Air Mobility (UAM) vehicles. The discussion is composed of the approach and progress toward a toolbox that integrates with existing NASA rotorcraft design software and processes. Several key challenges are highlighted including modeling requirements for failures, identification of critical failures, and capturing critical failures, with robustness to model uncertainty. A discussion of requirements for safety metrics, specific to UAM vehicles, as well as the effects of the control system design is also included. Results to date have demonstrated the degradation in flying qualities metrics due to propulsion failures.

George Altamirano↗
NASA NTRSDate not supplied

Enabling in-time Prognostics with Surrogate Modeling through Physics-enhanced Dynamic Mode Decomposition Method

Computational models provide essential quantitative tools for assessing and predicting the health and performance of physical systems. However, high-fidelity models are rarely used in real-time operations or large optimization loops, due to their time-intensive nature. A common approach to improving computational efficiency of prognosis is to employ surrogate models. Such models can significantly decrease computation time for some accuracy loss. In this context, use of Dynamic Mode Decomposition (DMD) is proposed to generate surrogate models for lithium-ion (Li-ion) battery discharge. DMD has been suggested and used successfully in the area of fluid dynamics for over a decade, but it has not been applied to the PHM domain, where far-ahead prediction of nonlinear behavior is crucial to propagate faults or predict Remaining Useful Life (RUL). For Li-ion battery health management, the standard application of DMD using only the observable quantities of interest was unable to capture the nonlinear discharge of batteries exhibited in lab testing. The Koopman theory, however, provides a mechanism to tradeoff low dimensional nonlinear models with high-dimensional linear ones in a DMD framework, by augmenting nonlinear state variables into the system representation. In this way, DMD allows for configurable simulation accuracy dependent on the dimensionality of the Koopman operator. For battery health management, we augmented the observable variables with the hidden states of a higher-fidelity physics model to build the DMD surrogate. In comparison to a high-fidelity model, the surrogate improved computational efficiency with only a minimal loss of accuracy, and enabled long-term prognostics horizons. A generalized method for this was implemented in the prog models python package.

prognostics and health management↗
NASA NTRSDate not supplied

Reasoning Service Exemplars for NASA’s Data and Reasoning Fabric

Future operations for Urban and Advanced Air Mobility are enabled by a distributed network of reliable and secured data and reasoning services referred to here as a fabric. In the aggregate, such a system must be all encompassing and mission agnostic, but specific use-cases are still needed to improve understanding and drive design paradigms. For this purpose, three reasoning service exemplars for Target Selection and Routing, Trajectory Generation, and Battery Health Management were developed and integrated into a specific NASA proposed data and reasoning fabric. These services were then used to build a mission reasoning application for lightning strike reconnaissance developed in collaboration with the Civil Air Patrol. Autonomous mission execution was then demonstrated using a multivehicle simulation platform with a full envelope 6 Degree-of-Freedom dynamics model for a concept electric Vertical Takeoff and Landing aircraft.

Autonomy↗
NASA NTRSDate not supplied

Flying Qualities Analysis and Piloted Simulation Testing of a Lift+Cruise Vehicle with Propulsion Failures in Hover and Low-Speed Conditions

The recent emergence of electric-Vertical Take-Off and Landing (eVTOL) vehicles for Urban Air Mobility (UAM) applications has resulted in a wide variety of configurations with unique stability and control characteristics. NASA is currently conducting research to develop conceptual design tools to accelerate public acceptance of these vehicles which includes requirements for safety during failure scenarios. This paper summarizes progress toward a toolbox for predicting flying qualities of eVTOL vehicles during critical propulsion failures that could impact the allowable design of the vehicle geometry or control system. Key topics include unique vulnerabilities of eVTOL/multirotor vehicles to propulsion failures, relevant flying qualities design metrics, and simulation modeling requirements for assessing flying qualities degradation due to failures. Results of a piloted simulation study conducted in the NASA Ames Vertical Motion Simulator (VMS) are presented. The VMS experiment was designed to assess and validate key handling qualities and safety design metrics for propulsion failures. These results show the correlation between control system design requirements and the degradation in handling qualities for various propulsion failures.

George Altamirano↗
NASA NTRSDate not supplied

Handling Qualities of Multirotor RPM-Controlled Electric-Vertical Take-Off and Landing (eVTOL) Aircraft for Urban Air Mobility (UAM)

A paradigm shift in rotorcraft design is being led by the prospect of propulsive forces being distributed across multiple rotors, such that each rotor can be directly driven by a dedicated electric motor. Crucially, some designers attempt to utilize these direct-drive mechanisms as the sole form of primary flight control. The feasibility of this design choice remains to be proven at the scales required for passenger transport. The paper presents a preliminary handling qualities analysis, for a six-passenger (1,200 lb payload) electric Hexacopter conceptual design, which shows that Level 1 handling qualities for limited agility operations are possible, provided that electric powertrains can deliver transient peak torques twice as high as the rated continuous torque of the conceptual design. Preliminary predictions are then substantiated by the results from a piloted handling qualities evaluation conducted in the NASA-Ames Vertical Motion Simulator (VMS). Three eVTOL configurations (a quadrotor, a hexacopter and a lift+cruise) with flight control laws implementing different levels of stability augmentation (Attitude Command-Attitude Hold and Translational Rate Command response types) were evaluated in four low speed and hover tasks requiring various levels of agility and precision.

UAM↗
NASA NTRSDate not supplied

Hybrid Modeling of Unmanned Aerial Vehicle Electric Powertrain for Fault Detection and Diagnostics

This paper shows the application of hybrid physics-informed machine learning to a representative electric powertrain for unmanned aerial vehicles. The model is composed of physics-derived principles and empirical equations, as well as fully connected networks that are strategically placed within the model to substitute equations that are subject to large uncertainty. Polynomial fit driven by heuristics or empirical observations can be substituted by more flexible networks that can minimize the error between model predictions and observations without being restricted to a predefined functional form. This modeling strategy allows training of networks deep inside the model and unknown parameters in a single learning stage. It has already been applied to Li-ion batteries in the past, and in this work, we extend the applications to other components of an electric powertrain, namely electronic speed controller with pulse-width modulation, and brushless DC motor with connected propeller. Training and testing of the model is carried out using experimental data from Li-ion battery discharge and powertrain testing in a laboratory environment.

Physics-Informed Machine Learning↗