Engineering Papers⌕ Search

Engineering topics

Garrett Sadler

Publications and source records attributed to Garrett Sadler.

At least 19 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS2024

m:N Working Group Meeting Summary November 2023

From November 28th to 30th, 2023 the m:N UAS working group and its subgroups [small Unmanned Aircraft Systems (sUAS), Large UAS, High Altitude Platform Systems (HAPS), and Urban Air Mobility (UAM)] met at the NASA Langley Research Center in Hampton, VA for an in person meeting. The subgroups meet multiple times throughout the year, virtually. Twice a year however, participants from all the subgroups come together in person to further identify and discuss challenges, and path forward ideas for incorporating UAS into the airspace. The m:N UAS working group is run by Jay Shively (Adaptive Aerospace) and Andy Thurling (Thurling Aero Consulting) and is comprised of members from government, industry, and academia in an effort to identify and reduce barriers to m:N operations. This effort also includes identifying requirements, use cases, and metrics to support organizations and groups including the FAA, RTCA, and ASTM. Each subgroup is run by a government/industry team (see below). sUAS Subgroup Garrett Sadler (NASA) Scott Scheff (HF Designworks) Large UAS Subgroup Conrad Rory (NASA) Brandon Suarez (Reliable Robotics) HAPS Subgroup Andy Thurling (Thurling Aero Consulting) Jeff Homola (NASA) UAM Subgroup Mike Politowicz (NASA) Scott Scheff (HF Designworks), member-at-large

m:N operations↗
NASA NTRS2023

m:N Working Group Annual Status Report

This document serves as an annual report of m:N Unmanned Aircraft Systems (UAS) subgroup activities, addressed challenges, and roadmaps for the future. The subgroups consist of small Unmanned Aircraft Systems (sUAS), Large UAS, High Altitude Platform Systems (HAPS), and Urban Air Mobility (UAM). Also included in this report are participant lists for each subgroup (Appendix B) and future roadmap and outreach plans.The subgroups meet in a virtual format multiple times throughout the year. Twice a year, participants from all the subgroups come together as part of the m:N working group to brief each other in person on progress, challenges, and path forward ideas for successful incorporation of UAS into the airspace. Recent m:N in person meetings include: - November 29-30, 2022 at the NASA Ames Research Center in Mountain View, CA - May 9 & 11, 2023 at AUVSI’s Xponential conference in Denver, CO - The next in person working group meeting is planned for November 28-30, 2023 at NASA Langley in Hampton, Virginia The m:N UAS working group is run by Jay Shively (Adaptive Aerospace) and Andy Thurling (Thurling Aero Consulting) and is comprised of members from government, industry, and academia in an effort to identify and reduce barriers to m:N operations. This effort also includes identifying requirements, use cases, and metrics to support organizations and groups including the Federal Aviation Administration (FAA) and Radio Technical Commission for Aeronautics (RTCA’s) SC-228 Detect and Avoid committee.

multi-vehicle↗
NASA NTRS2022

m:N Working Group

On November 29th and 30th, 2022, the m:N UAS working group and its subgroups [small Unmanned Aircraft Systems (sUAS), Large UAS, High Altitude Platform Systems (HAPS), and Urban Air Mobility (UAM)] met at the NASA Ames Research Center in Mountain View, CA for an in person meeting. The option to dial in remotely and use Conference.IO to engage with questions was offered as well. The subgroups meet multiple times throughout the year, virtually. Twice a year however, participants from all the subgroups come together to brief each other on progress, challenges, and path forward ideas for incorporating UAS into the airspace. The m:N UAS working group is run by Jay Shively (Adaptive Aerospace) and Andy Thurling (Thurling Aero Consulting) and is comprised of members from government, industry, and academia in an effort to identify and reduce barriers to m:N operations. This effort also includes identifying requirements, use cases, and metrics to support organizations and groups including the FAA and RTCA’s SC-228 Detect and Avoid. Each subgroup is run by a government/industry team.

multi-vehicle↗
NASA NTRS2020

A Human-in-theLoop Evaluation of ACAS Xu

As part of the Phase 2 UAS DAA MOPS, a Class 3 DAA system has been under development with the potential to resolve many of the limitations of Class 1 and 2 systems. A Class 3 system would combine the DAA and CA functions into a single, unified system, and would also extend the CA capabilities relative to TCAS II. Class 3 is enabled by the Airborne Collision Avoidance System (ACAS) XU, a next generation CA system developed specifically for UAS operations. Unlike Class 2 systems, ACAS XU provides CA protection against both cooperative and non-cooperative traffic. Class 3 systems also expand the CA logic to allow horizontal RAs in addition to vertical RAs. The purpose of the current study was to evaluate ACAS XU in a real-time, HITL simulation. The latest version of ACAS XU was implemented and pilots were tasked with responding to scripted traffic conflicts over the course of four experimental trials. Variables included the location of the ACAS XU guidance information (standalone vs. integrated) and traffic conflict type (DAA or CA threat). Sixteen active UAS pilots participated in the study, with ATC and ‘pseudo’ pilots acting as airspace confederates. Results showed that pilots were able to maintain DWC with ACAS XU at a rate comparable to previous research (~5%). Compliance rates to initial RAs were high (~90%) but dropped significantly when the target heading value issued during horizontal RAs were updated over the course of an encounter (30-70%). The implications of these findings on the display, alerting, and guidance requirements for Class 3 systems will be discussed.

unmanned aircraft systems↗
NASA NTRS2020

UAS Pilot Performance Comparisons with Different Low Size, Weight and Power Sensor Ranges

The present study evaluated the performance of UAS pilots under four simulated low size, weight, and power (SwaP) sensor ranges: 1.5nmi, 2.0nmi, 2.5nmi, and 3.0nmi. Nine active-duty UAS pilots responded to scripted DAA conflicts against non-cooperative intruders while flying a simulated RQ-7 Shadow at varied speeds along a pre-filed flight path in Class E airspace. Findings revealed a linear effect of sensor range on alerting time and separation performance, with nearly every DAA well clear (DWC) violation and all Near Mid-Air Collision (NMAC) events occurring below 2.5nmi. Response time differences at these reduced ranges were negligible due to the high frequency of warning-level alerts that require an immediate response. Since caution alert duration was truncated to some degree by each tested declaration range, pilots were often unable to coordinate their avoidance maneuvers with ATC prior to their uploads. Nonetheless, the 2.5nmi range allowed minimum alerting times that were sufficient for acceptable pilot performance. These findings will inform DAA system requirements for UAS with alternative surveillance equipment and aircraft performance capabilities. Implications on DAA display and sensor requirements are discussed.

unmanned aircraft systems↗
NASA NTRSDate not supplied

Display and Automation Considerations for the Airborne Collision Avoidance System Xu

In this presentation we examine several display and automation considerations of a collision avoidance system that is currently under development: the Airborne Collision Avoidance System (ACAS) Xu. This study builds on previous work conducted as part of NASA’s Unmanned Aircraft Systems (UAS) Integration into the National Airspace System (NAS) project. ACAS Xu represents the next-generation successor to the Traffic Alert and Collision Avoidance System (TCAS II), wherein the Xu variant is intended for UAS applications. Whereas TCAS II exclusively issues RAs in the vertical dimension, a major distinction between ACAS Xu and previous collision avoidance (CA) systems is the introduction of horizontal and “blended” RAs (i.e., RAs with both horizontal and vertical components). This present work was conducted as an engineering analysis involving two parts. In Part 1, a two-by-two, within-subjects study was performed that manipulated how RAs were presented to a pilot situated at a UAS ground control station. Five participants experienced four experimental trials in which text and aural alerting characteristics were manipulated. In Part 2, another five participants experienced four trials in which the levels of automation were manipulated with regard to the CA and return-to-course (RTC) tasks. The results for Part 1 found no effect of display or alerting configuration on pilot performance. However, it was discovered that pilot response time to RAs greatly depended on the RA type. In particular, pilots were quicker to respond to vertical RAs (M = 4.52 seconds) than horizontal (M = 7.42 seconds) and blended (M = 9.68 seconds) RAs in which both dimensions were issued simultaneously. For Part 2 of the study, pilots found both auto-CA and auto-RTC functions equally useful. Most pilots were comfortable with the automation, however responses were mixed. Three of five participants indicated high levels of comfort with the auto-CA function, while two rated their comfort as low. Pilots’ comfort for the auto-RTC functionality was slightly higher: four out of five pilots gave high ratings, while one pilot gave a low rating. Overall, pilots ordinally ranked their preference for automated functions as auto-CA together with auto-RTC (when an aural alert announces a change between CA and RTC states), auto-CA, and auto-CA and RTC (without the aural state-change announcement). Recommendations for improving the display of automation are also discussed.

collision avoidance↗
NASA NTRSDate not supplied

UAS Pilot Performance Comparisons with Different Low Size, Weight and Power Sensor Ranges

The present study evaluated the performance of UAS pilots under four simulated low size, weight, and power (SwaP) sensor ranges: 1.5nmi, 2.0nmi, 2.5nmi, and 3.0nmi. Nine active-duty UAS pilots responded to scripted DAA conflicts against non-cooperative intruders while flying a simulated RQ-7 Shadow at varied speeds along a pre-filed flight path in Class E airspace. Findings revealed a linear effect of sensor range on alerting time and separation performance, with nearly every DAA well clear (DWC) violation and all Near Mid-Air Collision (NMAC) events occurring below 2.5nmi. Response time differences at these reduced ranges were negligible due to the high frequency of warning-level alerts that require an immediate response. Since caution alert duration was truncated to some degree by each tested declaration range, pilots were often unable to coordinate their avoidance maneuvers with ATC prior to their uploads. Nonetheless, the 2.5nmi range allowed minimum alerting times that were sufficient for acceptable pilot performance. The study findings will inform DAA system requirements for UAS with alternative surveillance equipment and aircraft performance capabilities. Implications on DAA display and sensor requirements are discussed.

unmanned aircraft systems↗
NASA NTRSDate not supplied

A Human-in-the-Loop Evaluation of ACAS Xu

As part of the Phase 2 UAS DAA MOPS, a Class 3 DAA system has been under development with the potential to resolve many of the limitations of Class 1 and 2 systems. A Class 3 system would combine the DAA and CA functions into a single, unified system, and would also extend the CA capabilities relative to TCAS II. Class 3 is enabled by the Airborne Collision Avoidance System (ACAS) XU, a next generation CA system developed specifically for UAS operations. Unlike Class 2 systems, ACAS XU provides CA protection against both cooperative and non-cooperative traffic. Class 3 systems also expand the CA logic to allow horizontal RAs in addition to vertical RAs. The purpose of the current study was to evaluate ACAS XU in a real-time, HITL simulation. The latest version of ACAS XU was implemented and pilots were tasked with responding to scripted traffic conflicts over the course of four experimental trials. Variables included the location of the ACAS XU guidance information (standalone vs. integrated) and traffic conflict type (DAA or CA threat). Sixteen active UAS pilots participated in the study, with ATC and ‘pseudo’ pilots acting as airspace confederates. Results showed that pilots were able to maintain DWC with ACAS XU at a rate comparable to previous research (~5%). Compliance rates to initial RAs were high (~90%) but dropped significantly when the target heading value issued during horizontal RAs were updated over the course of an encounter (30-70%). The implications of these findings on the display, alerting, and guidance requirements for Class 3 systems will be discussed.

unmanned aircraft systems↗
NASA NTRSDate not supplied

A Human-in-the-Loop Evaluation of ACAS Xu

As part of the Phase 2 UAS DAA MOPS, a Class 3 DAA system has been under development with the potential to resolve many of the limitations of Class 1 and 2 systems. A Class 3 system would combine the DAA and CA functions into a single, unified system, and would also extend the CA capabilities relative to TCAS II. Class 3 is enabled by the Airborne Collision Avoidance System (ACAS) XU, a next generation CA system developed specifically for UAS operations. Unlike Class 2 systems, ACAS XU provides CA protection against both cooperative and non-cooperative traffic. Class 3 systems also expand the CA logic to allow horizontal RAs in addition to vertical RAs. The purpose of the current study was to evaluate ACAS XU in a real-time, HITL simulation. The latest version of ACAS XU was implemented and pilots were tasked with responding to scripted traffic conflicts over the course of four experimental trials. Variables included the location of the ACAS XU guidance information (standalone vs. integrated) and traffic conflict type (DAA or CA threat). Sixteen active UAS pilots participated in the study, with ATC and ‘pseudo’ pilots acting as airspace confederates. Results showed that pilots were able to maintain DWC with ACAS XU at a rate comparable to previous research (~5%). Compliance rates to initial RAs were high (~90%) but dropped significantly when the target heading value issued during horizontal RAs were updated over the course of an encounter (30-70%). The implications of these findings on the display, alerting, and guidance requirements for Class 3 systems will be discussed.

unmanned aircraft systems↗
NASA NTRSDate not supplied

UAS Pilot Performance Comparisons with Different Low Size, Weight and Power Sensor Ranges

The present study evaluated the performance of UAS pilots under four simulated low size, weight, and power (SwaP) sensor ranges: 1.5nmi, 2.0nmi, 2.5nmi, and 3.0nmi. Nine active-duty UAS pilots responded to scripted DAA conflicts against non-cooperative intruders while flying a simulated RQ-7 Shadow at varied speeds along a pre-filed flight path in Class E airspace. Findings revealed a linear effect of sensor range on alerting time and separation performance, with nearly every DAA well clear (DWC) violation and all Near Mid-Air Collision (NMAC) events occurring below 2.5nmi. Response time differences at these reduced ranges were negligible due to the high frequency of warning-level alerts that require an immediate response. Since caution alert duration was truncated to some degree by each tested declaration range, pilots were often unable to coordinate their avoidance maneuvers with ATC prior to their uploads. Nonetheless, the 2.5nmi range allowed minimum alerting times that were sufficient for acceptable pilot performance. These findings will inform DAA system requirements for UAS with alternative surveillance equipment and aircraft performance capabilities. Implications on DAA display and sensor requirements are discussed.

unmanned aircraft systems↗
NASA NTRSDate not supplied

M:N Operations NASA/Uber Collaboration

In this presentation, current approaches to enable multiple-operator, multiple vehicle (M:N) operations are reviewed together with recent collaborative efforts between NASA and Uber. Topics include a review of human-automation teaming (HAT) concepts, including plays and working agreements, and a particular task-allocation method called Automation Level-based Task Allocation (ALTA). Following introductory material on HAT, an overview of a recent (July 2020) cognitive walkthrough study of M:N operations in the context of a food delivery via small-Unmanned Aircraft Systems application is provided. Initial results from this cognitive walkthrough detailing operator feedback on displays, operator and supervisor roles and responsibilities, and the overall concept of operation are reviewed. The presentation concludes with a description of a future, human-in-the-loop simulation experiment of M:N operations in a high-fidelity environment, which will examine the effects of high workload and assistive automation/tools on operator performance.

human-automation teaming↗
NASA NTRSDate not supplied

m:N ConOps/R&R Remote Simulation

This presentation details the experimental design of an investigation of small unmanned aircraft system (sUAS) operations involving multiple vehicle management by a remote operator. The study is part of an ongoing effort to explore multiple vehicle control by multiple operators, i.e., the control of N vehicles by m operators (m:N operations). For this effort, NASA and collaborators have developed prototypes of a concept of operations (ConOps), roles and responsibilities (R&R) for operators and supervisors, and a ground control station (GCS), including software displays and interfaces. Participants in this study acted as the pilot-in-command of twelve aircraft flying pre-approved routes in a simulation of a food delivery operation utilizing sUAS in the San Diego, CA area. Each participant experienced four experimental trials. Twice within the course of each trial, participants were responsible for responding to a sudden, unanticipated, and high-priority contingency: an airspace restriction for sUAS operations known as a UAS Volume Reservation (UVR). Upon issuance of a UVR, pilots were expected to reroute affected vehicles around the airspace. The level of automation (LoA) and workload of the flight rerouting task were varied. The LoA was manipulated by providing reroute suggestions ("auto" condition) for aircraft or by requiring pilots to manually reroute ("manual" condition) affected vehicles. Workload was varied as a function of the number of vehicles affected by the UVR contingencies: 2 vehicles ("low workload" condition) versus 4 vehicles ("high workload" condition). Additionally, some vehicles required pilots to adjust for terrain conflicts while avoiding the UVR region. Due to the COVID-19 pandemic, in-person data collection for this study was not possible. Researchers adapted to this circumstance through the development of remote data collection protocol. Participants were able to view adapted GCS displays using the Microsoft Teams teleconferencing platform and responded to events by using a verbal protocol developed for the experiment. Using this protocol, participants provided instructions for actions to a researcher, referred to as the surrogate, to carry out on their behalf. This presentation describes the experiment design, including special details for remote data collection via a subject-surrogate configuration, and concludes with planned data analysis and results to be presented at a later date.

multi-UAS↗
NASA NTRSDate not supplied

m:N Working Group

The principles and reasons for employing Human Autonomy Teaming is discussed. Application of these techniques to multi-vehicle control is described. In several operational environments from small drone delivery to air taxi to autonomous cargo, many companies will need technologies that will allow for the operation of an unmanned aircraft (UAS or eVTOL) by a small crew of individuals that are located in a remote network operations center. That is, Multiple operators supervising an increasing Number of vehicles (M:N). This will require a new control/supervisory paradigm where the supervisors team with automation to achieve their joint tasks; Human Autonomy Teaming (HAT). This task will follow the HAT philosophy and tenants (e.g., trust, bi-directional communication, pilot directed interfaces). It will also develop and employ specific HAT tools (e.g., playbook, working agreements, predictive timeline displays, transparent interfaces to build trust). This presentation discusses the background and progress of the m:N working group.

multi-vehicle control↗
NASA NTRSDate not supplied

m:N Handoff Study

Explore the source record for details and available documents.

m:N operations↗
NASA NTRSDate not supplied

A New Control Paradigm: Multiple Aircraft Controlled by Multiple Operators

Remotely piloted aircraft systems (RPAS) are becoming more and more prevalent in the aerospace operations. This is true in a number of diverse domains; urban air mobility, medical product delivery, infrastructure inspection, high altitude pseudo-satellites, search and rescue, auto cargo and several other applications. One aspect that all of these share in common is the need for scalability to be viable and continue to grow. The Association of Uncrewed Vehicle Systems International (AUVSI) develops an annual economic report. They project that in the first three years of integration more than 70,000 jobs will be created in the US alone, with an economic impact of more than $13.6 billion. This benefit will grow through 2025 when we foresee more than 100,000 jobs created and economic impact of $82 billion. For many of these domains to reach these levels and have the scalability needed, they will require a remote pilot to control multiple aircraft (1:N) or the extension of that, multiple pilots controlling multiple aircraft (m:N). This is a new control paradigm that raises multiple issues in various areas. The issues include regulatory, technical, safety, community acceptance and Human Factors. Human factors issues include displays, pilot workload, pilot situation awareness just to name a few. This panel brings together researchers, developers and operators that have been working in the area of m:N. They will discuss the need, the issues and some potential solutions.

multi-vehicle control↗