Traditional air mobility capability development consists of defining a set of requirements and subsequently designing an aircraft that satisfies these requirements. Once designed, the overall structure or topology of the aircraft is fixed. That is, the aircraft has certain fixed physical dimensions, number and sizes of rotors, engines, etc. The topology cannot be changed to accomplish missions outside of its original design envelope, and if designed for a robust mission spectrum, the aircraft may be inefficient during routine missions that do not require peak performance. The work reported here takes a different approach where a core set of components (fuselage, rotor, power plant, etc) defines the fundamental modules of the air mobility system. The modules were developed using mature and readily available technologies and designed to be quickly connected to other modules to assemble different aircraft configurations with widely varying performance characteristics. The aircraft can be reconfigured between missions to best satisfy mission requirements. Thus, the envelope of achievable performance characteristics is significantly expanded compared to a single aircraft. This concept is explored for autonomous air cargo delivery missions targeted for nominal payloads and ranges around 500 pounds and 500 nautical miles with maximum payloads up to 3500 pounds across shorter distances. Results indicate that for mission spectrums with large differences between nominal and maximum payloads and skewed towards smaller payloads, the operational cost per lb-km is significantly reduced using a modular and reconfigurable air mobility system since cargo delivery requirements can be better matched to an air vehicle system.

Bleed-air control of ram-air parafoils is a lightweight and cost effective control mechanism that enables significant landing accuracy improvements compared to traditional trailing edge brake control. Single surface parafoil canopies have been shown to be lightweight canopies that possess similar flight behavior as their ram-air counterparts. The design of bleed air spoilers for single surface canopies is more challenging than for ram-air canopies because the internal ram air provides a high pressure differential at any chordwise location, unlike that of single surface canopies. This paper explores bleed-air spoilers on single surface parafoil canopies with a focus on turn rate and glide slope control capabilities. Through computational modeling and a flight test campaign, it is shown that bleed-air spoilers can provide both turn rate control from asymmetric vent opening and glide slope change from symmetric vent opening. Vents located at the 10% chord location yield substantial control response with a maximum turn rate response of 40 deg/sec and maximum glide slope change of 41%. Conversely, vents placed at 50% chord location yielded minimal turn rate control response.

Bleed-air control of ram-air parafoils is a lightweight and cost-effective mechanism that enables significant landing accuracy improvements compared to traditional trailing edge brake control due to the addition of direct glide slope control. Single-surface parafoil canopies have been shown to be lightweight and possess similar flight behavior as their ram-air counterparts. The design of bleed-air spoilers is more challenging for single-surface canopies, which lack the internal ram-air that provides a high pressure differential at any chordwise location. This paper explores bleed-air spoilers on single-surface parafoils with a focus on turn rate and glide slope control capabilities. Through a flight test campaign, it was shown that bleed-air spoilers on a single-surface parafoil provide both turn rate control from asymmetric vent opening and glide slope change from symmetric vent opening. Vents located far forward at the 10% chord location yield substantial control response with a maximum turn rate response of 38 deg/s and a maximum glide slope change of 58%. Vents located farther aft at 30, 50, and 70% chords demonstrated diminished control authority. Varying the spanwise locations of the spoilers revealed that the outermost vents had negligible lateral and longitudinal control authority.

After a sufficient amount of use under certain conditions, parachutes wear out and can no longer be safely employed. A simple and effective indicator of the remaining life of a parachute is the number of times it has been used, surviving opening shock, canopy inflation, transient parachute dynamics, and low velocity flight to the ground. This paper documents a small electronic device that automatically detects and logs each time a parachute has been used. Hardware and software that enables this device are described and data from bench and flight testing are detailed to highlight detection algorithm performance and long-term power consumption. Data shows that the proposed device accurately and reliably identifies individual parachute flights and that the ultra low power nature of the design permits the device to be continuously on for a period of time that is greater than the life span of the parachute. Thus, the device can be installed in the canopy during the fabrication process without the need for recharging or replacement of batteries.

Robotic landing gear (RLG) for rotorcraft improves performance in landing on sloped uneven terrain, unprepared areas, and ship decks. The interaction between the feet of the RLG and the landing surface are pivotal to a successful landing event. Slipping or bouncing of the feet can lead to a failed landing and a catastrophic accident. Proposed herein is the use of locking mechanisms on the RLG feet in order to eliminate landing gear slip and bounce during the landing event. Through the use of a comprehensive multibody dynamic simulation, locking mechanisms on the RLG feet are shown to eliminate landing event failures that can occur with nonlocking landing gear configurations, at the expense of a moderate increase in landing gear loads during a landing event. Results indicate that landing event failures are eliminated even in the situation where some feet-locking mechanisms are inoperable or break away. Furthermore, RLG with feet locking mechanisms permit the reduction or elimination of the need for active control of the RLG legs. The results herein give guidance to the development of integrated RLG with locking mechanisms.

Some air vehicles are configured such that major components exhibit relative motion with respect to one another where this relative motion significantly affects vehicle motion. In these cases, multibody flight dynamics is needed to adequately model system dynamics. This paper reports on a numerically efficient and versatile method for multibody flight dynamic simulation where the air vehicle is idealized as a collection of rigid bodies connected together by a set of joints. The method uses constrained coordinates with a constraint stabilization method based on a nonlinear control framework. The key innovation lies in relating the connections of the rigid bodies to an undirected graph and its adjacency matrix. By reordering connections, the bandwidth of the adjacency matrix can be minimized leading to substantial computational improvements. These improvements are applied to several air vehicle systems to highlight the computational benefits of the proposed technique.

The private eVTOL/UAM market is rapidly expanding, with many companies and investors betting on the possibility of low-cost-to-operate and zero emissions air taxis and last-mile-delivery sUAS dominating the skies of cities in the near future. While fixed wing and rotary wing aircraft regularly operate over populated areas, the envisioned orders-of-magnitude increase in air traffic increases the risk to those on the ground and in the air. The risk calculus is also altered by the novel designs of some eVTOL aircraft. Many designs have neither significant autorotation nor low-speed-gliding capability to land safely in the event of catastrophic power failure. Ballistic parachute systems (ballistic recovery systems), such as those found in increasing use on general aviation aircraft and small UAS (drones) provide some risk mitigation, but provide no control over the landing location; the aircraft will land wherever the wind blows it. Aviation safety experts have noted that while ballistic recovery systems improve safety and should be included on all UAM aircraft, there are still hazards from a vehicle descending uncontrolled under a parachute. This paper covers the analysis and design of a guided emergency recovery system, which uses bleed-air control and a gliding steerable parachute. The analysis shows that due to the use of a parachute which provides lift, rather than a pure drag device, and the use of lightweight bleed-air control, the overall weight of a guided recovery system can be lower than an equivalent ballistic recovery system. A key feature for an effective guided recovery system is selection of its landing point. The development of a massively parallel probabilistic evaluation method for selecting landing points which provide the best likelihood of avoiding hazards is shown with examples. Simulations of EGRES in New York City Urban Air Mobility, UAS interstate highway inspection, and UAS nuclear power plant inspection scenarios showed that EGRES increased the likelihood of landing in an acceptable area by 2.3, 1.8, and 2.9 times respectively, compared with a ballistic parachute recovery system. Finally, a preliminary design for a Group 2 UAS is presented, along with size and weight estimates.

This article examines the fundamental aspects of controlling ground resonance in rotorcraft equipped with actively controlled landing gear. Ground resonance is a mechanical instability affecting rotorcraft on the ground. It occurs at certain rotor speeds, where the lead–lag motion of the rotor couples with the motion of fuselage creating a self-excited oscillation. Typically, passive or semi-active lag dampers are used to avoid instability; however, these are undesirable from a design and maintenance perspective. Innovations in active landing gear for rotorcraft, such as articulated robotic legs, have provided an alternate approach to avoid the instability, eliminating the need for lag dampers with respect to ground resonance. This article extends classic ground resonance to include movable landing gear and identifies key physical parameters affecting dynamic behavior. Applying LQ optimal control to this model, it is shown that ground resonance instability can be eliminated using active landing gear as the control mechanism, even when there is no lag damping present in the rotor. In addition, while superior performance is achieved when landing gear movement can occur both longitudinally and laterally, it is still possible to stabilize ground resonance with inputs in a single direction, albeit with reduced performance.

There is a frequent need to cut lines or rigging on command in airdrop applications, and this is typically done with pyrotechnically driven cut knives. The use of hot wire cutters offers a simple, quiet, low-cost, low-weight alternative. This work provides an explanation of the principals behind hot wire cutters, outlines the design of an example hot-wire cutter system, provides results from experimental testing of this system on a variety of materials typically used in cargo airdrop, and shows results of the use of the system in small-scale airdrop testing. A simple, battery powered, small device was developed which can rapidly cut a wide variety of synthetic ropes including those made from nylon, polyester, Spectra, Dyneema, Vectran, and Kevlar. This device can be used to de-rig payloads from airdrop platforms, detach main parachutes from payloads after landing, and de-reef parachutes.

A fundamental consideration in the design of projectiles is the method of aerodynamic stabilization. The two categories of passive stabilization are fin stabilization and spin stabilization. Smart projectiles employ one of these two methods and enable active maneuvering with a physical control mechanism such as canards, pulse jets, microspoilers. This paper considers a moveable nose control mechanism that can be used for both stabilization and maneuvering. In particular, the paper focuses on the ability of a moveable nose to actively stabilize a nonspinning or slowly spinning projectile and permit the reduction or elimination of fins for stabilization. For a standard exemplar projectile, it is shown that an actively controlled nose is able to stabilize a nonspinning and finless projectile with reasonable control authority and associated bandwidth.

The public road setback distance is often an important factor that drives wind farm design. This paper outlines a methodology for assessing the risk imposed by blade throw at various road setbacks using a physics-based simulation approach. Given a road setback distance, Monte Carlo simulation is performed wherein blade throw parameters and vehicle locations are randomized. Potential collisions are determined using an “impact circle” approach which assumes that impact occurs if the vehicle is inside the impact radius of the blade fragment when it lands. This approach is exercised on several example turbines and risk levels are calculated for various road setbacks. The method is also applied to a notional wind farm with turbines located at a typical road setback distance. Results show that the blade throw risk imposed to vehicles on public roads for the example wind farm is extremely small and commensurate with risks imposed by everyday activities.

Guided airdrop systems offer the ability to deliver payloads with accuracy and with a high payload to air vehicle weight fraction. Traditionally, these systems use airborne guidance units with trailing edge control to steer ram-air parafoils in the same manner as a human pilot. Trailing edge control only provides lateral (turn rate) and limited airspeed control. Earthly Dynamics, LLC developed an in-canopy precision guided aerial delivery system that offers both lateral steering and longitudinal glide slope control of a ram-air parafoil. This paper will present the feasible mission CONOPS for this technology as a cargo delivery, spacecraft recovery, or personnel autopilot system. This work then details the flight-proven hardware and software architecture implemented in a 116 ft$^2$ ram-air parachute. Scalability of the design up to 2400 lb payloads is presented. Data and configuration for multiple flight test demonstrations as a cargo airdrop system and a spacecraft recovery system are provided. Finally, this paper concludes with future applications and development plans for this technology.

Impact attenuation is a critical component of cargo airdrop systems, ensuring payload survivability while allowing practical impact speeds. Cardboard honeycomb is the standard material used for this purpose, but this material is far too stiff to function well on small (5 – 50) lb payloads. This work describes an effort to investigate alternative impact attenuation options for small, lightweight payloads. A purpose-built impact test rig was constructed to facilitate rapid testing and enable repeatable impacts at various combinations of horizontal and vertical impact speeds. A computer vision system was developed to track the payloads as they impact to measure actual impact speeds, and custom sensors were used to measure impact accelerations. A variety of different impact attenuation materials were tested using the test rig and instrumentation systems and the results of these experiments are reported.

Robotic technology has given rise to many advances in aviation as well as shaped the design of air vehicles. The goal of this work is to synthesize the recent advances in vehicle design as it pertains to aerial robotics. A secondary use of this work is for guidance on the different vehicle design options one has in the space of aerial robotics. Within aerial robotics, vehicle design and morphology are diversifying to meet the needs of expanding mission profiles. Robotics technology, such as robotic legs and arms, are integrating on to aerial systems with improved results. Aerial robotic missions have grown in diversity, and choice of vehicle to accomplish a mission is important. There are many different aerial robotic platforms: rotorcraft, fixed-wing, rocketry, flapping-wing, canopy, lighter-than-air, and hybrid designs. There are also many robotic features that can be integrated into an aerial system: robotic landing gear, aerial manipulators, and interfacing.

Aero-driven bleed-air (ADB) is a flow control technology that offers a unique alternative to novel aircraft concepts that are not conducive to traditional flapped control surfaces. The primary source of mechanical energy for ADB control is derived from the motion of the vehicle. Inherent pressure differences across aerodynamic surfaces are communicated by means of internal channels, with the intent of affecting the local crossflow. Effected across a large area, ADB can significantly alter the global aerodynamic forces and moments. This investigation explores the performance of large-area-distributed ADB as a roll control mechanism on a low-speed, fixed-wing air vehicle. A series of wind tunnel experiments is presented, aimed at characterizing the effects that various porosity configurations have on the lift and drag of a Clark-Y airfoil section. ADB was found to be capable of reducing the sectional lift coefficient across a broad domain of angles of attack (-8 degrees<alpha<28 degrees) by as much as Delta cl=-0.7relative to the baseline airfoil. A mathematical model for estimating the steady-state roll rate of a wing with ADB is developed. Several pore opening schedules are considered, with an emphasis on finding a nearly linear relationship between bleed level and steady-state roll rate. The selected schedule is implemented on a flight test vehicle equipped with ADB roll control effectors. Roll rates of up to100 deg/swere achieved in flight, and flight test data were found to validate the steady-state roll rate model. The aircraft was piloted from takeoff through landing using ADB for roll control.

In this work a novel crashworthy cable-driven four-bar link mechanism is designed, manufactured, and tested for the development of robotic landing gear for rotorcraft. The paper will demonstrate how combining two such mechanisms through a common driving cable leads to unique advantages in the kinematics of the system and reduced loads applied to the actuators driving the system. The results on the kinematics and static force distributions are presented for such a configuration and show that, for the rotorcraft class under consideration (220-440 lb, 100-200 kg), the leg mechanism can conform to ground slopes up to similar to 20 deg. A novel design, manufacturing, and a drop testing workflow that allows for rapid design and iteration of robotic landing gear are further developed. The workflow includes the design, simulation, rapid manufacturing, and crash-survivability testing through drop testing with a two week turnaround time for a full design iteration. With regard to rapid manufacturing, this work demonstrates the use of three-dimensionally printed acrylonitrile butadiene styrene mandrels for the production of fully functional carbon fiber polymer reinforced components. This paper concludes with actuation experiments to show practical performance capabilities of the final crashworthy robotic landing gear system designed.

Guided airdrop systems have traditionally used position and velocity information from a GPS receiver as their only source of feedback. The use of additional sensors in the guidance units is challenging because sensors in the guidance unit are in close proximity to powerful electric motors. Furthermore, there is a large amount of relative motion between the guidance unit and the parachute as they are coupled by a flexible network of rigging lines. By placing sensors in the parachute itself, it is possible to obtain accurate estimates of the canopy motion and orientation with low-cost sensors requiring minimal calibration. Specialized in-canopy sensor pods were developed to provide distributed sensing throughout a parachute canopy and a sensor fusion algorithm was developed to combine the raw data from these sensor pods into useful canopy state estimates. The effectiveness of this approach is demonstrated first in simulation, and then with flight test results on full-scale airdrop systems. The rich feedback signal available from in-canopy sensors can provide improved datasets for more detailed system identification as well as enabling novel guidance, navigation and control approaches which will lead directly to improved landing accuracy.

Parameter estimation and system identification of smart projectiles is an important and commonly used industrial tool. Existing methods rely on good initial estimates of the parameters to avoid local minima and ensure proper convergence. New projectile configurations may include highly nonlinear dynamics or unknown control parameters that cannot be known a priori. Anew method for projectile parameter estimation is proposed that combines an output error parameter estimation algorithm with meta-optimization. Meta-optimization uses a suite of optimizers in an intelligent manner to reliably minimize a cost function. This new method is applied to the identification of a smart projectile system equipped with microspoilers using simulated spark range data. The method is able to reliably estimate the aerodynamic coefficients of the projectile body as well as the properties of the control mechanism based on a fit of multiple trajectories.