Search NASA⌕ Search

SEARCH · Search NASA

Results for “RL”

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.

188 records · Page 11
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

NASA's Space Launch System Begins Integration, Stacking in Preparation for Artemis I Launch

The Artemis era of human lunar exploration is nearing take-off as NASA’s new super heavy-lift launch vehicle, the Space Launch System (SLS), begins stack-ing and integration operations in mid-2020 at Kennedy Space Center (KSC) in Florida. With a planned upgrade path to progressively more powerful vehicles and availability in crew and cargo configurations, SLS provides a unique and flexible launch solution to send crew, large-scale infrastructure and robotic probes to deep space. The SLS Block 1 vehicle, the initial variant to fly, is optimized for lunar missions with a proven propulsion system consisting of four liquid hydrogen (LH2)/liquid oxygen (LOX)-fed RS-25 engines and twin five-segment solid rocket boosters (SRBs). The Block 1 vehicle can also be outfitted with an industry-standard 5 m-class payload fairing (the “cargo” configuration) and will launch at least 27 metric tons (t) of mass to trans-lunar injection (TLI). SLS is the backbone of NASA’s Artemis program, which will return the agency’s human spaceflight program to the Moon for the first time since 1972. For the Artemis I mission, SLS will send an uncrewed Orion spacecraft to TLI, where it will enter a distant retrograde lunar orbit and fly 38,000 nmi past the Moon – farther than any spacecraft built for humans has ever traveled. The SLS Block 1 vehicle for Artemis I completed manufacturing in 2019. Several elements, including the upper stage, have been delivered to the Exploration Ground Systems (EGS) program at KSC and are being prepped for integration and stack-ing. The five-segment solid rocket boosters – the largest and most powerful ever built for flight – are also complete. The booster motor segments for the Artemis I flight are scheduled to ship from prime contractor Northrop Grumman’s Utah facilities and begin stacking and integration at KSC in June 2020. The SLS core stage is the largest rocket stage NASA has ever built in terms of volume and height, and includes the avionics and the tanks that feed cryogenic propellant to the four RS-25s (formerly Space Shuttle Main Engines [SSMEs]). They have been modified with an updated controller and nozzle insulation to protect them from the hotter launch environment. The SLS core stage is currently being test-ed at NASA’s Stennis Space Center (SSC) in a series of “green run” tests to verify it meets design and performance requirements. Following the green run test series, which is scheduled to culminate with a full-duration hot-fire of the four RS-25 engines, the core stage will ship to KSC and be stacked between the sol-id rocket boosters in the Vehicle Assembly Building (VAB). Integration of the vehicle will continue with the upper stage, known as the Interim Cryogenic Propulsion Stage (ICPS) and the Launch Vehicle Stage Adapter (LVSA) on the core stage. Another adapter, the Orion Stage Adapter (OSA), connects SLS to Orion and provides housing for 13 6U CubeSat payloads manifested on Artemis I. The CubeSats will be released in deep space after Orion separates from the vehicle, and the flight marks the first ride share opportunity for independent small-sats to deep space. The second major SLS variant to come online, Block 1B, replaces the single-engine ICPS with a four-engine LH2/LOX Exploration Upper Stage (EUS). This more powerful upper stage, along with other vehicle up-grades, will enable the Block 1B vehicle to launch 38-42 t to TLI, depending on crew or cargo configuration. The final evolution of the vehicle, Block 2, will onramp evolved solid rocket boosters to increase mass to TLI to 43-46 t, de-pending on crew or cargo configuration. The Block 1B/Block 2 vehicles can be outfitted with an 8.4 m-diameter payload fairing in 19.1 m or 27.4 m lengths, to provide unprecedented volume for payloads. With the initial Block 1 vehicle completely manufactured and the core stage in final testing before shipping to KSC, the SLS Program and its industry partners have made significant progress manufacturing subsequent vehicles. For the second Block 1 vehicle, the solid rocket motor segments are complete, as are the RS-25 engines with controllers. All five major components of the Artemis II core stage – the forward skirt, LOX and LH2 tanks, intertank and engine section – are manufactured and technicians are installing subsystems at NASA’s rocket factory, Michoud Assembly Facility. The RL-10 engine for the Artemis II ICPS is complete and panels have been machined for its LH2 tank. In addition, panels are machined for the vehicle’s two adapters, with welding scheduled to begin in summer 2020. Flight hard-ware is also in production for the third SLS vehicle, with several booster motor segments cast. The pace of development on the EUS has increased, with the goal to complete Critical Design Review (CDR) in December 2020. Several EUS test rings have been machined at Michoud. The EUS is designed to exe-cute a variety of missions – human spaceflight, deployment of deep-space infra-structure, or high-C3 missions to the outer solar system – with crew and cargo configurations available beginning in the mid-2020s. The near-term goal for the nation’s powerful new space exploration asset, however, is to launch the Arte-mis program, and send the first woman and the next man to the lunar surface. At the Astrodynamics Specialist Conference, the SLS program will update the community on the progress of the initial Block 1 vehicle in final green run test-ing, integration and stacking. In addition, this paper will provide an update to the community on the manufacturing status of subsequent Block 1 and Block 1B vehicles.

Steve Creech↗
NASA NTRS2023

Reinforcement Learning Approach to Flight Control Allocation with Distributed Electric Propulsion

The flight control system of the SUSAN Electrofan concept aircraft achieves attitude control using both conventional flight control surfaces and differential thrust through distributed electric propulsion (DEP) from sixteen wing-mounted electric engines. The introduction of eight pairs of wing fans for attitude control creates a highly actuated system. Such a system requires more sophisticated control to operate, especially in the presence of wingfan failures where the loss of a single wingfan can result in a thrust imbalance. This paper investigates the use of deep reinforcement learning (RL) using proximal policy optimization (PPO) to achieve attitude control through a combination of DEP and control surface deflections. First, the paper examines the aircraft undergoing a coordinated turn. Then, it examines the aircraft experiencing a wingfan failure during cruise conditions. It is shown that deep reinforcement learning can be a potential avenue for nonlinear flight control design.

Distributed Electric Propulsion↗
DOE OSTI2026

Learning to Trigger: Reinforcement Learning at the Large Hadron Collider

High-throughput scientific facilities such as the Large Hadron Collider depend on real-time event filtering (\textit{triggering}) under tight constraints on bandwidth, latency, and storage. In practice, trigger menus are largely static and hand-tuned and can become suboptimal as detector conditions, pileup, and background composition drift over time. We cast online threshold tuning as a sequential decision-making problem: a reinforcement learning agent ingests streaming summaries of recent rates and signal-sensitive features and updates trigger thresholds to maximize signal efficiency while tracking a target background rate within a tolerance band. We adapt Group-Filtered Policy Optimization (GFPO) to streaming control and introduce two variants (GFPO-F, GFPO-FR) that enforce background rate feasibility during training. On a benchmark that emulates realistic collider operation, we study two representative triggers: a total transverse energy ($H_{T}$) trigger sensitive to pileup variation, and an anomaly-detection (AD) trigger based on reconstruction loss for rare or non-standard signatures. On Monte Carlo streams, our agent increases the fraction of in-tolerance time intervals by 48% ($H_T$) and 28% (AD), with a cumulative gain of up to 2% in signal efficiency on those in-tolerance intervals. Transferring from simulation to \emph{real} collision data (CMS Run 283408), the same agent, without fine-tuning, achieves a 56% ($H_T$) and 28% (AD) in-tolerance improvement over baselines, with further signal-efficiency gain on both triggers. To our knowledge, this is the \emph{first} demonstration of RL-based trigger control on real Large Hadron Collider collision data. Code is available at https://github.com/Zixind/GFPO_LHC (see repo for details).

Ding, Zixin [Chicago U.]↗
DOE OSTI2026

Intern Poster Session 08/13: Autonomous Nuclear Robotics: Applications in nuclear waste inspection and hot cell experiments

The nuclear industry is experiencing renewed interest in autonomous robotics, yet most deployed systems remain teleoperated with limited autonomy. This work presents two contributions toward fully autonomous nuclear robotic systems: autonomous waste inspection at the Hanford Site and an autonomous hot cell laboratory framework. Inspections of Hanford's underground waste storage tanks are performed manually at significant cost and personnel exposure. We developed a reinforcement-learning (RL) training pipeline for a custom-built inspection arm. In parallel, we are designing an autonomous laboratory framework for post-irradiation examination in hot cells at the Specimen Preparation Laboratory (SPL) that integrates computer vision, task and motion planning, hardware execution, and operator-in-the-loop control. These systems demonstrate a path toward safer, more efficient nuclear operations by reducing human exposure while maintaining rigorous human oversight at critical decision points.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

From Sim to Real: A Pipeline for Training and Deploying Traffic Smoothing Cruise Controllers

Designing and validating controllers for connected and automated vehicles to enhance traffic flow presents significant challenges, from the complexity of replicating real-world stop-and-go traffic dynamics in simulation, to the intricacies involved in transitioning from simulation to actual deployment. In this work, we present a full pipeline from data collection to controller deployment. Specifically, we collect 772 km of driving data from the I-24 in Tennessee, and use it to build a one-lane simulator, placing simulated vehicles behind real-world trajectories. Using policy-gradient methods with an asymmetric critic, we improve fuel efficiency by over 10% when simulating congested scenarios. Our comprehensive approach includes reinforcement learning for controller training, software verification, hardware validation and setup, and navigating various sim-to-real challenges. Furthermore, we analyze the controller's behavior and wave-smoothing properties, and deploy it on four Toyota Rav4’s in a real-world validation experiment on the I-24. Lastly, we release the driving dataset, the simulator and the trained controller, to enable future benchmarking and controller design.

42 ENGINEERING↗
NASA NTRS2021

Reconciling Carbon-cycle Processes from Ecosystem to Global Scales

Understanding carbon (C) dynamics from ecosystem to global scales remains a challenge. Although expansion of global carbon dioxide (CO2) observatories makes it possible to estimate C-cycle processes from ecosystem to global scales, these estimates do not necessarily agree. At the continental US scale, only 5% of C fixed through photosynthesis remains as net ecosystem exchange (NEE), but ecosystem measurements indicate that only 2% of fixed C remains in grasslands, whereas as much as 30% remains in needleleaf forests. The wet and warm Southeast has the highest gross primary productivity and the relatively wet and cool Midwest has the highest NEE, indicating important spatial mismatches. Newly available satellite and atmospheric data can be combined in innovative ways to identify potential C loss pathways to reconcile these spatial mismatches. Independent datasets compiled from terrestrial and aquatic environments can now be combined to advance C-cycle science across the land–water interface.

carbon cycle↗
NASA NTRS2024

MSFC-TVC | TB-03: Derivation of Thrust Vector Control (TVC) Actuator-Force / Gimbal-Torque Transformation Matrix

Thrust vector control (TVC) systems for rocket engine propulsion traditionally use a simple linear relationship to convert between actuator forces and torques about the engine gimbal’s center-of-rotation (COR). As shown in Equation (1), the torque about the gimbal COR is proportional to the applied actuator force and the TVC moment arm (MA)—the perpendicular distance between the TVC actuator’s line-of-action (LOA) and the engine gimbal’s COR. While this fundamental relationship remains valid and accurate in a two-dimensional (2D), one-degree-of-freedom (1-DOF) context—particularly in its non-linear formulation as described in the ER63 Technical Bulletin TB-02 (Derivation of Thrust Vector Control (TVC) Engine-Gimbal / Actuator Moment-Arm Geometry)—it becomes limited when extended to three-dimensional (3D), two-degree-of-freedom (2-DOF) analyses. In 3D, out-of-plane angular displacements can arise, causing the gimbaled engine plane-of-motion to become non-coplanar with the respective actuator plane-of-motion. Such misalignments occur due to the rod-end (RE) and/or tail-stock (TE) mounting geometry of individual TVC actuators. These geometric complexities lead to inaccuracies in calculating engine torques and corresponding actuator forces if only the traditional TVC MA relationship is employed. For higher gimbal angular displacements (e.g., greater than 5 degrees), these inaccuracies become more pronounced, necessitating a more robust mathematical framework. This document presents a comprehensive 3D (2-DOF) derivation of a transformation matrix that accurately converts between gimbaled engine torques and TVC actuator forces. By incorporating the necessary geometric and rotational considerations, this new approach corrects the limitations of the traditional TVC MA method. Subsequent sections compare the newly formulated approach to the traditional method, demonstrating its enhanced accuracy and reliability for a broad range of gimbaled engine conditions.

Thrust Vector Control↗