Search NASA⌕ Search

SEARCH · Search NASA

Results for “Trajectory Design”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,621 records · Page 90

HARMONY: Large-Scale Architecture Search for Efficient Hybrid Language Models

As large language models scale to trillions of parameters, their computational and memory requirements present critical challenges for efficient training and deployment. While Mixture of Experts (MoE) architectures enable efficient scaling through sparse parameter activation, and state-space models like Mamba offer linear-time complexity, principled methods for combining these paradigms remain undeveloped. We introduce HARMONY (Hybrid Architecture Research for Mamba, Optimized with Neural efficiencY), a multi-objective evolutionary neural architecture search framework for discovering efficient hybrid language models that integrate Transformer attention mechanisms, Mixture-of-Experts routing, and Mamba state-space components. Through large-scale distributed search using 16,384 MI250X GPUs on the Frontier supercomputer, HARMONY explores a comprehensive design space encompassing six attention variants (MHA, MQA, GQA, MLA, SWA, and Mamba-2), variable MoE configurations with both routed and shared experts, and extensive Mamba hyperparameters. Our framework discovers heterogeneous architectures that balance training performance with computational efficiency through multi-objective optimization incorporating latency penalties and fitness-based selection. Analysis of discovered architectures reveals that optimal hybrid designs favor heterogeneous component mixing rather than homogeneous patterns, with Mamba-2 and Multi-Head Latent Attention (MLA) emerging as preferred mechanisms. Discovered architectures demonstrate superior training efficiency: our best configuration achieves a final perplexity of 1.0874 with 2.38B parameters while processing 4,320 tokens/second, outperforming significantly larger manually designed models. Full-scale evaluation shows HARMONY's top architectures achieve better loss trajectories than equivalently-sized models using state-of-the-art configurations including Mixtral, Jamba, and Samba. Additionally, we demonstrate 91% weak scaling efficiency when training discovered 36B-parameter models across 1,024 GPUs. HARMONY is released as an open framework with comprehensive tools for building and training hybrid models using expert-data-pipeline parallelism, democratizing access to automated architecture design for next-generation language models.

Herron, Emily [ORNL] (ORCID:0000000273008172)↗

Wind Field and Trajectory Models for Tornado-Propelled Objects

A mathematical model to predict the trajectory of tornado born objects postulated to be in the vicinity of nuclear power plants is developed. An improved tornado wind field model satisfied the no slip ground boundary condition of fluid mechanics and includes the functional dependence of eddy viscosity with altitude. Subscale wind tunnel data are obtained for all of the missiles currently specified for nuclear plant design. Confirmatory full-scale data are obtained for a 12 inch pipe and automobile. The original six degree of freedom trajectory model is modified to include the improved wind field and increased capability as to body shapes and inertial characteristics that can be handled. The improved trajectory model is used to calculate maximum credible speeds, which for all of the heavy missiles are considerably less than those currently specified for design. Equivalent coefficients for use in three degree of freedom models are developed and the sensitivity of range and speed to various trajectory parameters for the 12 inch diameter pipe are examined.

Redmann, G. H.↗

Wake Fields in the Super B Factory Interaction Region

The geometry of storage ring collider interaction regions present an impedance to beam fields resulting in the generation of additional electromagnetic fields (higher order modes or wake fields) which affect the beam energy and trajectory. These affects are computed for the Super B interaction region by evaluating longitudinal loss factors and averaged transverse kicks for short range wake fields. Results indicate at least a factor of 2 lower wake field power generation in comparison with the interaction region geometry of the PEP-II B-factory collider. Wake field reduction is a consderation in the Super B design. Transverse kicks are consistent with an attractive potential from the crotch nearest the beam trajectory. The longitudinal loss factor scales as the -2.5 power of the bunch length. A factor of 60 loss factor reduction is possible with crotch geometry based on an intersecting tubes model.

43 PARTICLE ACCELERATORS↗

Determination of Ceres Physical Parameters Using Radiometric and Optical Data

The Dawn spacecraft was launched on September 27th, 2007. Its mission is to rendezvous with and observe the two largest bodies in the main asteroid belt, Vesta and Ceres. It has completed over a year’s worth of direct observations of Vesta from early 2011 through late 2012. In the spring of 2015, the Dawn spacecraft entered orbit around the asteroid Ceres for the start of what is expected to be more than a year of science operations. The science data collected from this encounter consist of infrared (IR) images and spectra, visible images through a number of color filters, gamma ray detections and measurements of the Ceres gravity field. These data will be collected during several science phases: an Approach phase (1500000-4860 km from Ceres), a Survey orbit (4860 km radius), a High Altitude Mapping Orbit (HAMO) (1940 km radius) and a Low Altitude Mapping Orbit (LAMO) (855 km radius). The Approach phase included three Rotational Characterization (RC) opportunities. Designing each science orbit and successfully transferring into that orbit requires a sufficiently accurate estimate of Ceres physical parameters (body fixed frame, GM and harmonics). This paper focuses on work performed to estimate Ceres physical parameters using Deep Space Network (DSN) radiometric tracking data and optical measurements derived from science camera imagery. This paper describes planning for the data acquisition, as well as processing techniques and methodology. The trajectories predicted by the gravity field estimations are also compared with the actual as-flown trajectories. Observations of the gravity at high altitudes are found to be sufficient to design precision orbits at lower altitudes. Follow-up analysis after successfully reaching LAMO is included, as is a discussion of lessons learned.

Kennedy, Brian M.↗

The vibro-acoustic mapping of low gravity trajectories on a Learjet aircraft

Terrestrial low gravity research techniques have been employed to gain a more thorough understanding of basic science and technology concepts. One technique frequently used involves flying parabolic trajectories aboard the NASA Lewis Research Center Learjet aircraft. A measurement program was developed to support an isolation system conceptual design. This program primarily was intended to measure time correlated high frequency accelerations (up to 100 Hz) present at various locations throughout the Learjet during a series of trajectories and flights. As suspected, the measurements obtained revealed that the environment aboard such an aircraft can not simply be described in terms of the static level low gravity g vector obtained, but that it also must account for both rigid body and high frequency vibro-acoustic dynamics.

Grodsinsky, C. M.↗

Orion Entry Display Feeder and Interactions with the Entry Monitor System

The Orion spacecraft is designed to return astronauts to a landing within 10 km of the intended landing target from low Earth orbit, lunar direct-entry, and lunar skip-entry trajectories. Al pile the landing is nominally controlled autonomously, the crew can fly precision entries manually in the event of an anomaly. The onboard entry displays will be used by the crew to monitor and manually fly the entry, descent, and landing, while the Entry Monitor System (EMS) will be used to monitor the health and status of the onboard guidance and the trajectory. The entry displays are driven by the entry display feeder, part of the Entry Monitor System (EMS). The entry re-targeting module, also part of the EMS, provides all the data required to generate the capability footprint of the vehicle at any point in the trajectory, which is shown on the Primary Flight Display (PFD). It also provides caution and warning data and recommends the safest possible re-designated landing site when the nominal landing site is no longer within the capability of the vehicle. The PFD and the EMS allow the crew to manually fly an entry trajectory profile from entry interface until parachute deploy having the flexibility to manually steer the vehicle to a selected landing site that best satisfies the priorities of the crew. The entry display feeder provides data from the ENIS and other components of the GNC flight software to the displays at the proper rate and in the proper units. It also performs calculations that are specific to the entry displays and which are not made in any other component of the flight software. In some instances, it performs calculations identical to those performed by the onboard primary guidance algorithm to protect against a guidance system failure. These functions and the interactions between the entry display feeder and the other components of the EMS are described.

Baird, Darren↗

Kinetic Monte Carlo Simulations of Chemical Reactions on Surfaces

Lattice based kinetic Monte Carlo (kMC) simulations are a powerful tool for micro-kinetic modeling. The kinetic Monte Carlo technique stochastically simulates chemical reaction processes (e.g. adsorption, desorption, diffusion), where the probability of a given reaction process being selected is weighted by its rate constant. The open source software package, KMOS1, allows fast construction of kMC simulations by automatically implementing a computationally efficient kMC framework, thus allowing the researcher to focus on accurate parameterization and formulation of the kinetic model. From this, the spatiotemporal evolution (e.g. concentration, surface configuration, turnover frequencies) of crystal surfaces or bulk crystalline structures is obtained. KMOS uses the variable step-size method implementation of kMC. At each step, the VSSM updates available events, calculates the corresponding transition rates and simulation time, and selects one of the available events. The lattice based kMC technique has many potential applications within SRNL where chemical kinetics are of interest. By transitioning through time along a trajectory of configuration space, fundamental insights to a system's dominant reactions and parameters can be gained, leading to improved materials and process design. In this work, we have established capabilities for fast implementation of kMC simulations through the installation and testing of the KMOS software on SRNL high performance computing resources. The parameterization of kMC simulations is often performed from first-principles energetic calculations. For the purposes of this work, the test cases use arbitrary species and thus, have arbitrary parameterization. In order to demonstrate the use of KMOS, two primary test cases are shown. Case 1 involves simple adsorption and desorption. Case 2 involves a slightly more complex mechanism. The evolution of the surface concentration and the final surface configuration for simulation Case 2 is shown. Rate-limiting processes and rate-limiting configurations are not always known or understood, but kMC provides information pertaining to both. The capability to implement efficient micro-kinetic models using the kinetic Monte Carlo technique has been established through the testing and installation of KMOS on SRNL HPC resources.

97 MATHEMATICS AND COMPUTING↗

Flight Test Results: CTAS Cruise/Descent Trajectory Prediction Accuracy for En route ATC Advisories

The Center/TRACON Automation System (CTAS), under development at NASA Ames Research Center, is designed to assist controllers with the management and control of air traffic transitioning to/from congested airspace. This paper focuses on the transition from the en route environment, to high-density terminal airspace, under a time-based arrival-metering constraint. Two flight tests were conducted at the Denver Air Route Traffic Control Center (ARTCC) to study trajectory-prediction accuracy, the key to accurate Decision Support Tool advisories such as conflict detection/resolution and fuel-efficient metering conformance. In collaboration with NASA Langley Research Center, these test were part of an overall effort to research systems and procedures for the integration of CTAS and flight management systems (FMS). The Langley Transport Systems Research Vehicle Boeing 737 airplane flew a combined total of 58 cruise-arrival trajectory runs while following CTAS clearance advisories. Actual trajectories of the airplane were compared to CTAS and FMS predictions to measure trajectory-prediction accuracy and identify the primary sources of error for both. The research airplane was used to evaluate several levels of cockpit automation ranging from conventional avionics to a performance-based vertical navigation (VNAV) FMS. Trajectory prediction accuracy was analyzed with respect to both ARTCC radar tracking and GPS-based aircraft measurements. This paper presents detailed results describing the trajectory accuracy and error sources. Although differences were found in both accuracy and error sources, CTAS accuracy was comparable to the FMS in terms of both meter-fix arrival-time performance (in support of metering) and 4D-trajectory prediction (key to conflict prediction). Overall arrival time errors (mean plus standard deviation) were measured to be approximately 24 seconds during the first flight test (23 runs) and 15 seconds during the second flight test (25 runs). The major source of error during these tests was found to be the predicted winds aloft used by CTAS. Position and velocity estimates of the airplane provided to CTAS by the ATC Host radar tracker were found to be a relatively insignificant error source for the trajectory conditions evaluated. Airplane performance modeling errors within CTAS were found to not significantly affect arrival time errors when the constrained descent procedures were used. The most significant effect related to the flight guidance was observed to be the cross-track and turn-overshoot errors associated with conventional VOR guidance. Lateral navigation (LNAV) guidance significantly reduced both the cross-track and turn-overshoot error. Pilot procedures and VNAV guidance were found to significantly reduce the vertical profile errors associated with atmospheric and aircraft performance model errors.

Green, S.↗

Sensitivity of the Asteroid Redirect Robotic Mission (ARRM) to Launch Date and Asteroid Stay Time

National Aeronautics and Space Administrations (NASAs) proposed Asteroid Redirect Mission (ARM) is being designed to robotically capture and then redirect an asteroidal boulder mass into a stable orbit in the vicinity of the moon, where astronauts would be able to visit and study it. The current reference trajectory for the robotic portion, ARRM, assumes a launch on a Delta IV H in the end of the calendar year 2021, with a return for astronaut operations in cislunar space in 2026. The current baseline design allocates 245 days of stay time at the asteroid for operations and boulder collection. This paper outlines analysis completed by the ARRM mission design team to understand the sensitivity of the reference trajectory to launch date and asteroid stay time.

trajectory analysis↗

Aerothermal Design of a Common Probe for Multiple Planetary Destinations

Estimate the mass of the Thermal Protection System (TPS) for a single design construct of an atmospheric entry probe with a rigid aeroshell, which could be used at five destinations, i.e. Venus, Saturn, Uranus, Neptune, and perhaps, Jupiter. The entry mass of the probe is 400 kg with a ballistic coefficient of 216 kg/m2. Process: The 3DoF trajectory simulation program Traj, coupled with the TPS response program FIAT was used for simulation and design. The assumed atmospheric models were VIRA (Venus-GRAM) for Venus, the Julianne Moses' model for Saturn, a NASA Ames engineering model for Uranus, Neptune-GRAM for Neptune, and Galileo Probe (Al Seiff's) result for Jupiter.

Allen, Gary A., Jr.↗

Pluto flyby mission design concepts for very small and moderate spacecraft

This paper discusses the trade-offs involved in the design of a Pluto flyby mission using a low-mass spacecraft, the objective of which is the first-ever reconnaissance of the Pluto/Charon system. The analyses include launch opportunity/trajectory identification/launch vehicle trades, flight time vs net spacecraft mass trades, and flyby targeting considerations. Two mission concepts currently under development are discussed. One concept involves an 83 kg dry spacecraft which uses a direct trajectory and a Titan IV/Centaur (with two solid rocket motor upper stages) to reach Pluto in 7 years. The other design is a 316 kg dry spacecraft which uses a 3+Delta VEJGA trajectory and a Delta II (7925) to arrive at Pluto in 13.6 years.

Weinstein, S. S.↗

Helicopter flight control system design and evaluation for NOE operations using controller inversion techniques

A flight control system design technique is described that allows rapid evaluation of candidate control laws using computer simulated trajectories. The two-step technique first computes the optimum trajectory for the basic aircraft, and then reconstructs the pilot's control displacements necessary to fly this trajectory, given a particular control system architecture. Since the flight control system itself does not add to the acceleration potential of the helicopter, one may evaluate several proposed control laws in terms of the resulting stick inputs for the same maneuver. The method is illustrated through application to implicit and explicit model-following control laws designed for a battlefield helicopter with and without an auxiliary propulsor. It is shown that the technique is a valuable aid for selection of promising control laws for further simulator studies and flight investigations.

Mckillip, Robert M., Jr.↗

Aerocapture Design Study for a Titan Polar Orbiter

In 2014 a team at NASA Goddard Space Flight Center (GSFC) studied the feasibility of using active aerocapture to reduce the chemical Delta V requirements for inserting a small scientific satellite into Titan polar orbit. The scientific goals of the mission would be multi-spectral imaging and active radar mapping of Titan's surface and subsurface. The study objectives were to: (i) identify and select from launch window opportunities and refine the trajectory to Titan; (ii) study the aerocapture flight path and refine the entry corridor; (iii) design a carrier spacecraft and systems architecture; (iv) develop a scientific and engineering plan for the orbital portion of the mission. Study results include: (i) a launch in October 2021 on an Atlas V vehicle, using gravity assists from Earth and Venus to arrive at Titan in January 2031; (ii) initial aerocapture via an 8-km wide entry corridor to reach an initial 350X6000 km orbit, followed by aerobraking to reach a 350X1500 km orbit, and a periapse raise maneuver to reach a final 1500 km circular orbit; (iii) a three-part spacecraft system consisting of a cruise stage, radiator module, and orbiter inside a heat shield; (iv) a 22-month mission including station keeping to prevent orbital decay due to Saturn perturbations, with 240 Gb of compressed data returned. High-level issues identified include: (i) downlink capability - realistic downlink rates preclude the desired multi-spectral, global coverage of Titan's surface; (ii) power - demise of the NASA ASRG (Advanced Stirling Radioisotope Generator) program, and limited availability at present of MMRTGs (Multi-Mission Radioisotope Generators) needed for competed outer planet missions; (iii) thermal - external radiators must be carried to remove 4 kW of waste heat from MMRTGs inside the aeroshell, requiring heat pipes that pass through the aeroshell lid, compromising shielding ability; (iv) optical navigation to reach the entry corridor; (v) the NASA requirement of continuous critical event coverage for the orbiter, especially during the peak heating of the aerocapture when the radio link will be broken. In conclusion, although Titan aerocapture allows for considerable savings in propellant mass, this comes at a cost of increased mission complexity. Further architecture study and refinement is required to reduce high-level mission risks and to elucidate the optimum architecture.

Nixon, Conor A.↗

Trajectory Reconstruction of Orion Ground Test Article Water Impact Tests

The Orion Crew Module is a component of NASA’s Multi-Purpose Crew Vehicle that will be used for future missions to low Earth orbit and beyond. Ten water impact tests of the Orion Ground Test Article (GTA) were conducted at the Hydro Impact Basin at NASA Langley Research Center in 2016 and were designed to provide data for the validation of the LS-DYNA model used to determine the Crew Module structural loads during ocean splashdown, and the determination of an acceptable Model Uncertainty Factor to apply to simulation results used to drive the design. Post-test data obtained from the onboard sensors were used to reconstruct the GTA trajectories both before and after water impact. Results from one vertical test and two swing tests are presented and compared to videos taken for each test.

Lugo, Rafael A.↗

Mars InSight Entry, Descent, and Landing Trajectory and Atmosphere Reconstruction

The InSight mission landed on the surface of Mars on November 26th, 2018. The InSight system performance met all design requirements, although several performance metrics fell near the boundaries of the predictions. The peak deceleration was high, the overall timeline was short, and the landing site was uprange and crossrange from the target. This paper describes the reconstruction of the entry, descent, and landing trajectory and atmosphere. The approach utilizes a Kalman filter to blend sensor data to obtain the vehicle trajectory. The aerodynamic database is used in combination with the sensed accelerations to obtain estimates of the atmosphere-relative state, which in turn is used to derive the free-stream atmospheric conditions during entry, until the time of parachute deployment. The results indicate that the reconstructed atmosphere was approximately 1σbelow the preflight atmosphere. Analysis of the reconstructed vehicle attitude angles indicate that the aerodynamic lift was oriented downward at entry. The vehicle developed a roll rate during entry, which directed a component of the lift to the north. The low density and aerodynamic lift direction are determined to be the primary causes of the high deceleration, short timeline, and location of the landing site relative to the target.

Christopher D Karlgaard↗

Mars InSight Entry, Descent, and Landing Trajectory and Atmosphere Reconstruction

The InSight mission landed on the surface of Mars on November 26th, 2018. The InSight system performance met all design requirements, although several performance metrics fell near the boundaries of the predictions. The peak deceleration was high, the overall timeline was short, and the landing site was uprange and crossrange from the target. This paper describes the reconstruction of the entry, descent, and landing trajectory and atmosphere. The approach utilizes a Kalman filter to blend sensor data to obtain the vehicle trajectory. The aerodynamic database is used in combination with the sensed accelerations to obtain estimates of the atmosphere-relative state, which in turn is used to derive the free-stream atmospheric conditions during entry, until the time of parachute deployment. The results indicate that the reconstructed atmosphere was approximately 1σbelow the preflight atmosphere. Analysis of the reconstructed vehicle attitude angles indicate that the aerodynamic lift was oriented downward at entry. The vehicle developed a roll rate during entry, which directed a component of the lift to the north. The low density and aerodynamic lift direction are determined to be the primary causes of the high deceleration, short timeline, and location of the landing site relative to the target.

Christopher D Karlgaard↗

Design of Electrostatic Dust Lofting Suborbital Flight Experiment Examining Photoionization under Lunar Gravity

Dust on the lunar surface electrostatically charges due to the plasma environment surrounding the Moon, causing grains to become lofted and adhere to nearby surfaces including landers and astronauts. Studying the behaviors of these charged particles in the lunar environment is essential to plan around the deleterious effects of dust to future Moon missions. Models attempt to predict the amount of dust loading that can be expected in many of these scenarios, but they require experimental validation to be predictive. This physics cannot be fully studied on Earth due to the six times larger gravitational force obscuring the electrostatic interactions, so it is necessary to run experiments in a more relevant environment, including vacuum and near-lunar gravitational effects. An experiment has been designed to fly on the Lunar Gravity Acceleration (LGA) mission aboard the Blue Origin New Shepard suborbital rocket. This experiment will perform photoionization charging of lunar regolith simulant grains under the illumination of an ultraviolet (UV) source. As a result, the charged grains will then electrostatically repel one another and loft in the reduced gravity environment; their trajectories will be imaged via a high-speed camera. Preliminary laboratory results influencing the design of this experiment will be presented, including characterization of several UV sources, measurements of photoionization currents under various vacuum conditions, and examination of lunar simulant dust lofting under terrestrial gravity. Results from this flight will be compared with ground-based testing and the laboratory results outlined above to examine the dependence on gravity and will be fed into the dust charging and lofting models currently under development.

electrostatics↗

VIPRE: A Tool Aiding the Design for Entry Probe Missions

Exploring planetary atmospheres uncovers important information for how our solar system formed and evolved. While remote sensing is extensively used, some crucial observations require in-situ measurements by an atmospheric probe. Given their scientific importance, probe missions to Saturn, Uranus and Neptune are considered for the coming decades. In anticipation of future probe missions, the software tool VIPRE was developed as proof-of-concept to facilitate selection of probe entry locations. Currently, there is no analytical way to identify which interplanetary trajectory from thousands of feasible launch opportunities is optimal for a considered mission concept. The search and decision process for that solution is complex and relies on the intuition of mission designers, who focus on a subset of trajectories to make the trade space manageable. The idea of VIPRE is to (1) generate a multi-dimensional data cube showing relevant engineering and science parameters simultaneously for thousands of trajectories, and (2) visualize the data for all entry sites over the body's envelope. VIPRE lays the foundation to make the data available for browsing in a 3-D visualization to identify the best family of solutions for a given mission. The paper introduces the validated and verified core algorithms of VIPRE, published on GitHub. VIPRE serves as a basic framework to be used and extended for different purposes. The paper presents the motivation for the development and algorithms. It explains the computation and data visualization strategy, and gives a list of suggested functionalities to extend and further develop VIPRE to fully leverage its potential.

Ice Giants↗