Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “online adaptivity”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
NASA NTRS2024

An On-Board Off-Board Framework for Online Replanning: Applied to UAVs in Urban Environments

Autonomous systems are being used in a multitude of areas at an increasing rate and require a high level of adaptivity and intelligence to operate safely, especially under faulty conditions. This paper introduces a novel genetic algorithm tailored for UAV trajectory replanning, with an improved execution time via search space reduction based on the operating conditions of the UAV and its remaining mission. A unique characteristic of the replanning agent is its fast-start and adaptive properties, pre-seeding candidates with partial solutions and dynamically tuning elitism, crossover, and mutation rates in correspondence to the average fitness and diversity of the population. A population restart mechanism and early stopping mechanism are evaluated as well to assess their effect on solution quality and runtime. Previous work on genetic algorithms for UAV replanning were conducted with short trajectories in a small state space. Our UAV operates in a 56,000 square meter simulated urban environment, with static obstacles and a total of 53 possible waypoints. The agent increases the safety and reliability of UAV autonomy when operating under faulty conditions and when replanning is required.

Machine Learning↗
DOE OSTI2022

Online Learning and Pricing for Service Systems with Reusable Resources

We consider a price-based revenue management problem with finite reusable resources over a finite time horizon T. Customers arrive following a price-dependent Poisson process, and each customer requests one unit of c homogeneous reusable resources. If there is an available unit, the customer gets served within a price-dependent exponentially distributed service time; otherwise, the customer waits in a queue until the next available unit. In this paper, we assume that the firm does not know how the arrival and service rates depend on posted prices, and thus it makes adaptive pricing decisions in each period based only on past observations to maximize the cumulative revenue. Given a discrete price set with cardinality P, we propose two online learning algorithms, termed batch upper confidence bound (BUCB) and batch Thompson sampling (BTS), and prove that the cumulative regret upper bound is O˜(√PT) , which matches the regret lower bound. In establishing the regret, we bound the transient system performance upon price changes via a novel coupling argument, and also generalize bandits to accommodate subexponential rewards. Here, we also extend our approach to models with balking and reneging customers and discuss a continuous price setting. Our numerical experiments demonstrate the efficacy of the proposed BUCB and BTS algorithms.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1985

Integrated Procurement Management System, Version II

Integrated Procurement Management System, Version II (IPMS II) is online/ batch system for collecting developing, managing and disseminating procurementrelated data at NASA Johnson Space Center. Portions of IPMS II adaptable to other procurement situations.

Collier, L. J.↗
DOE OSTI2025

Online Bayesian State Estimation for Real-Time Monitoring of Growth Kinetics in Thin Film Synthesis

Rapid validation of newly predicted materials through autonomous synthesis requires real-time adaptive control methods that exploit physics knowledge, a capability that is lacking in most systems. Here, in this study, we demonstrate an approach to enable real-time control of thin film synthesis by combining in situ optical diagnostics with a Bayesian state estimation method. We developed a physical model for film growth and applied the direct filter (DF) method for real-time estimation of nucleation and growth rates during pulsed laser deposition (PLD). We validated the approach using simulated and experimental reflectivity data for WSe 2 growth and ultimately deployed the algorithm on an autonomous PLD system during the growth of 1T'-MoTe 2 . The DF robustly estimates growth parameters in real time at early stages of growth, down to 15% monolayer area coverage. This fusion of in situ diagnostics, data assimilation, and physical modeling opens new opportunities in adaptive control of synthesis trajectories toward desired material states.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Online energy consumption forecast for battery electric buses using a learning-free algebraic method

Accurately predicting the energy consumption plays a vital role in battery electric buses (BEBs) route planning and deployment. Based on the algebraic derivative estimation, we present a novel method to forecast the energy consumption in real time. In contrast to the mainstream machine-learning-based methods, the proposed method does not require access to the historical energy consumption data. It eliminates the time-consuming and computationally expensive offline training. Consequently, its prediction performance is not constrained by the quantity and quality of the training data. Moreover, the method can swiftly adapt to new situations not included in the previous driving cycles, which makes it especially suitable for emerging transport modes, e.g., on-demand transit services. In addition, its online execution only involves algebraic calculations, yielding superior calculation efficiency. Using real-world data, we comprehensively compare the performance of the proposed learning-free algebraic method with multiple representative machine-learning-based methods. Finally, the advantages and limitations of the proposed method are discussed in detail.

33 ADVANCED PROPULSION SYSTEMS↗
NASA NTRS2017

Overview: Performance Adaptive Aeroelastic Wing

An overview of recent aeroelasitc wing-shaping work at the NASA Ames Research Center is presented. The highlight focuses on activity related to the Performance Adaptive Aeroelastic Wing concept and related Variable Camber Continuous Trailing Edge Flap actuation system. Topics covered include drag-reducing configurations and online algorithms, gust and maneuver load techniques, and wind tunnel demonstrations.

Hashemi, Kelley↗
DOE OSTI2025

Virtual to Physical: Reinforcement Learning to Optimize SNS Particle Accelerator Controls

Complex accelerators must have control systems that can handle dynamic nonlinear environments. This makes traditional control methods unsuitable as they can struggle to adapt to these uncertainties. This provides an ideal environment for reinforcement learning algorithms as they are adaptable and generalizable. We present a reinforcement learning pipeline that can effectively handle the dynamics of a complex accelerator. We test and prove our pipelines capabilities on multiple environments including the Spallation Neutron Source (SNS) and the Beam Test Facility (BTF) at Oakridge National Lab (ORNL). Due to the limited time available to train an online algorithm like reinforcement learning on a real accelerator, we utilize a virtual twin accelerator (VIRAC) developed by ORNL to pretrain the policy and show its ability to converge in the virtual environment. We then test the adaptability of the pretrained RL model by applying it on the real accelerator and comparing the results. Utilizing our Scientific Optimization and Controls Toolkit (SOCT) and open-source standards such as Gymnasium we create and solve for a MEBT orbit correction problem in the SNS and an emittance maximization problem in the BTF. We show how Twin Delayed Deep Deterministic Policy Gradient (TD3) can solve this optimization environment in the virtual accelerator and transfer this policy onto the real accelerator for inference and model retraining. We show how reinforcement learning can be utilized as a control system for complex accelerators and provide a model pipeline for how an implementation performs and can be adapted to new accelerator control problems.

Kasparian, Armen [Thomas Jefferson National Accele↗
DOE OSTI2023

An Online Prototype Toolset for Predicting and Optimizing P&T Performance (FY23 Status Report)

A new web-based toolset is being developed to support ongoing remediation optimization efforts and implementation of an adaptive site management strategy for the 200 West Area Pump-and-Treat (P&T) system at the Hanford Site. This toolset, comprising the well performance index tool and the well optimization pre-screening tool, will offer a user-friendly interface to predict and optimize the P&T well network’s performance at a preliminary level. Efforts in fiscal year (FY) 2023 focused on three main components: updating the existing deep learning model for predicting P&T performance, designing and developing a prototype of a web-based performance index tool, and initiating the conceptual design of the well optimization pre-screening tool. The well performance index tool is based on a pre-trained deep learning model that allows users to select a target contaminant and well screen length, then visualize the predicted performance of potential new wells across the site. The well optimization pre-screening tool includes two separate modules: the pre-computed scenario viewer, which organizes and visualizes offline optimization simulation results, and the quick analysis module, which provides real-time model prediction using user-specified well locations. In FY24, the plan is to add web-based applications to SOCRATES for both the well performance prediction tool and the optimization prescreening tool, with accompanying user and theory guides. These tools are intended to enable an accessible, easily applied, and transparent approach to remedy planning and decision-making.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2011

Feature Acquisition with Imbalanced Training Data

This work considers cost-sensitive feature acquisition that attempts to classify a candidate datapoint from incomplete information. In this task, an agent acquires features of the datapoint using one or more costly diagnostic tests, and eventually ascribes a classification label. A cost function describes both the penalties for feature acquisition, as well as misclassification errors. A common solution is a Cost Sensitive Decision Tree (CSDT), a branching sequence of tests with features acquired at interior decision points and class assignment at the leaves. CSDT's can incorporate a wide range of diagnostic tests and can reflect arbitrary cost structures. They are particularly useful for online applications due to their low computational overhead. In this innovation, CSDT's are applied to cost-sensitive feature acquisition where the goal is to recognize very rare or unique phenomena in real time. Example applications from this domain include four areas. In stream processing, one seeks unique events in a real time data stream that is too large to store. In fault protection, a system must adapt quickly to react to anticipated errors by triggering repair activities or follow- up diagnostics. With real-time sensor networks, one seeks to classify unique, new events as they occur. With observational sciences, a new generation of instrumentation seeks unique events through online analysis of large observational datasets. This work presents a solution based on transfer learning principles that permits principled CSDT learning while exploiting any prior knowledge of the designer to correct both between-class and withinclass imbalance. Training examples are adaptively reweighted based on a decomposition of the data attributes. The result is a new, nonparametric representation that matches the anticipated attribute distribution for the target events.

Thompson, David R.↗
DOE OSTI2023

Online transfer learning strategy for enhancing the scalability and deployment of deep reinforcement learning control in smart buildings

In recent years, advanced control strategies based on Deep Reinforcement Learning (DRL) proved to be effective in optimizing the management of integrated energy systems in buildings, reducing energy costs and improving indoor comfort conditions when compared to traditional reactive controllers. However, the scalability and implementation of DRL controllers are still limited since they require a considerable amount of time before converging to a near-optimal solution. This issue is currently addressed in literature through the offline pre-training of the DRL agent. However this solution results in two main critical issues: (1) the need to develop a building surrogate model to perform the training task, and (2) the need to perform a fine-tuning process over several training episodes to obtain a near-optimal control policy. In this context, this paper introduces an Online Transfer Learning (OTL) strategy that exploits two knowledge-sharing techniques, weight-initialization and imitation learning, to transfer a DRL control policy from a source office building to various target buildings in a simulation environment coupling EnergyPlus and Python. A DRL controller based on discrete Soft Actor–Critic (SAC) is trained on the source building to manage the operation of a cooling system consisting of a chiller and a thermal storage. Several target buildings are defined to benchmark the performance of the OTL strategy with that of a Rule-Based Controller (RBC) and two DRL-based control strategies, deployed in offline and online fashion. The strategy adopted for OTL emulates the real world implementation with a simulation process by implementing the transferred DRL agent for a single episode in the target buildings. Target buildings have the same geometrical features and are served by the same energy system as the source building, but differ in terms of weather conditions, electricity price schedules, occupancy patterns, and building envelope efficiency levels. The results show that the OTL strategy can reduce the cumulated sum of temperature violations on average by 50% and 80% respectively when compared to RBC and online DRL while enhancing the energy system operation with electricity cost savings ranging between 20% and 40%. Furthermore, the OTL agent performs slightly worse than the offline DRL controller but it does not require any modeling effort and can be implemented directly on target buildings emulating a real-world implementation.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2022

Multi-modal Energy-optimal Trip Scheduling in Real-time (METS-R) for Transportation Hubs (Final Report)

This report summarizes the work performed under the award number EE0008524. The project develops the Multi-modal Energy-optimal Trip Scheduling in Real-time (METS-R) platform as the next-generation transportation solution based on autonomous electric vehicles (AEV) serving passenger trips from and to urban transportation hubs, to substantially reduce transportation energy consumption. Extensive data collection and analyses were first conducted to understand the demand patterns and energy consumption of hub-based on-road trips. Then, a data-driven framework that consists of an analytical module and a simulation module was proposed. For the analytical module, five planning + operation tools were developed to support the planning and energy-efficient operations of urban AEV services: the charging station planning that robotically allocates charging supplies based on the stationary charging demand distribution; the transit planning and demand adaptive scheduling model that efficiently generates\ candidate transit routes from hubs to other places and dynamically adjusts the transit time table to fit the current demand; the online energy-efficient routing that learns the energy-optimal paths from observations of link-level energy consumption in real-time; the hub-based ridesharing that matches trip requests together with account for the uncertainty of future trip demand and vehicle supply; and finally, the integrated demand prediction and anomaly detection pipeline that leverages the flight/train time table and support other planning/operation tools. To demonstrate the performance of these tools, a scalable high-performance agent-based simulator was built. We divided the urban space into multiple service zones where each zone was considered as an agent for passenger generation and vehicle charging. Two types of AEV agents were coded to model two types of mobility services: AEV taxi and AEV transit. For the AEV taxi, the team implemented the functions of pickup/drop-off passengers, energy-efficient routing, ridesharing, fleet rebalancing, and recharging. For the AEV bus, the team implemented the functions of demand-adaptive route scheduling, passenger boarding, and recharging. A high-performance computing framework was introduced to receive various profiling information (such as link energy updates, vehicle speed) from the simulator instances and communicate the operational commands back to the instances. The numerical experiments show that each of the proposed operational algorithms can reduce energy consumption and improve system efficiency. Furthermore, there exists the need to collectively consider multiple planning + operational strategies as multiple strategies can influence each other in terms of performance impacts. Recommendations for future work related to AEV planning and simulation are discussed.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
NASA NTRS2017

Synergies of Subaru and CGI

Given the limited observing time and demanding scenarios of the WFIRST coronagraph instrument (CGI), it is critical to consider how Subaru observations can benefit its observing program. Subaru telescope has a suite of instruments with their adaptive optics (AO) and extreme adaptive optics modules (SCExAO). With SCExAO, the Subaru telescope is capable of detection and spectral characterization of binaries and bright (greater than 5(exp -6) contrast) companions in the near-infrared. This will enable the vetting of targets, disk detection and characterization, and potentially some additional science should CGI identify interesting targets during its technology demonstration and potential guest observer program. Additionally, large companions that are within the inner working angle of the coronagraph can be identified using the VAMPIRES aperture masking interferometer. With highly complementary target brightness and significantly overlapping fields of view, there is a great deal of potential for combined observations with Subaru and CGI. This will represent the first time single observations spanning the visible to near-infrared will be possible for high contrast imaging. We will discuss the overlap of instrumentation over time, the implication of instrument evolution as TMT comes online, and how this can be used to improve both science and technology demonstrations for CGI.

adaptive optics (AO↗
NASA NTRS1986

Knowledge-based load leveling and task allocation in human-machine systems

Conventional human-machine systems use task allocation policies which are based on the premise of a flexible human operator. This individual is most often required to compensate for and augment the capabilities of the machine. The development of artificial intelligence and improved technologies have allowed for a wider range of task allocation strategies. In response to these issues a Knowledge Based Adaptive Mechanism (KBAM) is proposed for assigning tasks to human and machine in real time, using a load leveling policy. This mechanism employs an online workload assessment and compensation system which is responsive to variations in load through an intelligent interface. This interface consists of a loading strategy reasoner which has access to information about the current status of the human-machine system as well as a database of admissible human/machine loading strategies. Difficulties standing in the way of successful implementation of the load leveling strategy are examined.

Chignell, M. H.↗
NASA NTRS2014

Forecasting Sensorimotor Adaptability from Baseline Inter-Trial Correlations

One of the greatest challenges surrounding adaptation to the spaceflight environment is the large variability in symptoms, and corresponding functional impairments, from one crewmember to the next. This renders preflight training and countermeasure development difficult, as a "one-size-fits-all" approach is inappropriate. Therefore, it would be highly advantageous to know ahead of time which crewmembers might have more difficulty adjusting to the novel g-levels inherent to spaceflight. Such knowledge could guide individually customized countermeasures, which would enable more efficient use of crew time, both preflight and inflight, and provide better outcomes. The primary goal of this project is to look for a baseline performance metric that can forecast sensorimotor adaptability without exposure to an adaptive stimulus. We propose a novel hypothesis that considers baseline inter-trial correlations, the trial-to-trial fluctuations in motor performance, as a predictor of individual sensorimotor adaptive capabilities. To-date, a strong relationship has been found between baseline inter-trial correlations and adaptability in two oculomotor systems. For this project, we will explore an analogous predictive mechanism in the locomotion system. METHODS: Baseline Inter-trial Correlations: Inter-trial correlations specify the relationships among repeated trials of a given task that transpire as a consequence of correcting for previous performance errors over multiple timescales. We can quantify the strength of inter-trial correlations by measuring the decay of the autocorrelation function (ACF), which describes how rapidly information from past trials is "forgotten." Processes whose ACFs decay more slowly exhibit longer-term inter-trial correlations (longer memory processes), while processes whose ACFs decay more rapidly exhibit shorterterm inter-trial correlations (shorter memory processes). Longer-term correlations reflect low-frequency activity, which is more easily measured in the frequency domain. Therefore, we use the power spectrum (PS), which is the Fourier transform of the ACF, to describe our inter-trial correlations. The decay of the PS yields a straight line on a log-log frequency plot, which we quantify by Beta = - (slope of PS on log-log axes). Hence, Beta is a measure of the strength of inter- trial correlations in the baseline data. Larger Beta values are indicative of longer inter-trial correlations. Experimental Approach: We will begin by performing a retrospective analysis of treadmill-gait adaptation data previously collected by Dr. Bloomberg and colleagues. Specifically, we will quantify the strength of inter-trial correlations in the baseline step cadence and heart rate data and compare it to the locomotor adaptability performance results already described by these investigators. Incorporating these datasets will also allow us to explore the applicability of (and potential limitations surrounding) the use of Beta in forecasting physiological performance. We will also perform a new experiment, in which Beta will be derived from baseline data collected during over-ground (non-treadmill) walking, which will enable us to consider locomotor performance, through the parameter Beta, under the most functionallyrelevant, natural gait condition. This experiment will incorporate two baseline and five post-training over-ground locomotion tests to explore the consistency and potential adaptability of the Beta values themselves. HYPOTHESES: We hypothesize that the strength of baseline inter-trial correlations of step cadence and heart rate will relate to locomotor adaptability. Specifically, we anticipate that individuals who show weaker longer-term inter-trial correlations in baseline step cadence data will be the better adaptors, as step cadence can be modified in real-time (i.e., online corrections are an inherent property of the locomotor system; analogous to results observed in the VOR). Conversely, because heart rate is not altered mid-beat, we expect that individuals who demonstrate stronger longer-term correlations in heart rate will be the better adaptors (analogous to results observed in the saccadic system). CONCLUSIONS: At the conclusion of this project we hope to uncover a baseline predictor of locomotor adaptability. If our hypotheses hold true, our results will demonstrate that the temporal structure of baseline behavioral data contains important information that may aid in forecasting adaptive capacities. The ability to predict such adaptability in the sensorimotor system has significant implications for spaceflight, where astronauts must adjust their motor programs following a change in g-level to retain movement accuracy.

Beaton, K. H.↗
NASA NTRSDate not supplied

Ensemble Forecasting/Assimilation/Emissions Estimation with WRF-Chem/DART: Accomplishments, Lessons Learned, and Future Plans

Over the past ten years, we have been conducting research on regional ensemble atmospheric composition forecasting/data assimilation/emissions estimation with WRF-Chem/DART. WRF- Chem/DART integrates the Weather Research and Forecasting model (WRF) with online chemistry (WRF-Chem) into the Data Assimilation Research Testbed (DART). DART is an ensemble data assimilation system based on the ensemble adjustment Kalman filter (EAKF) with adaptive inflation, localization (physical and state space), and an optional non-Gaussian formulation of the EAKF. DART includes assimilation of meteorological and limited chemical observations. WRF-Chem/DART extends DART to include assimilation of: MOPITT CO; IASI CO and O3; MODIS AOD; OMI O3, NO2, and SO2; TROPOMI CO, O3, NO2, and SO2, TES CO, CO2 (research mode), O3, NH3, and CH4 (research mode); CrIS CO, O3, NH3, CH4 (research mode), and PAN; SCIAMACHY NO2; GOME2a NO2; MLS O3 and HNO3; and proxy TEMPO O3, and NO2 satellite retrievals as raw retrievals or as ‘compact phase space retrievals’ (CPSRs) for profile retrievals. WRF-Chem/DART also assimilates in situ atmospheric composition measurements and uses the ‘state augmentation method’ for emissions estimation. In our presentation, we will provide an overview of WRF-Chem/DART: • applications and results; • lessons learned from: (i) independent versus joint assimilation; (ii) total/partial column versus profile retrieval assimilation; (iii) joint in situ and retrieval assimilation; (iv) assimilation at grid resolutions ranging from 100 km to 4 km; (v) dynamic emissions estimation; and • future work related to wildfire emissions estimation and intercomparison of CMAQ (online)/JEDI and CMAQ (online and offline)/DART.

WRF-Chem/DART↗
DOE OSTI2026

FLYING SERVING: On-the-Fly Parallelism Switching for Large Language Model Serving

Production LLM serving must simultaneously deliver high throughput, low latency, and sufficient context capacity under non-stationary traffic and mixed request requirements. Data parallelism (DP) maximizes throughput by running independent replicas, while tensor parallelism (TP) reduces per-request latency and pools memory for long-context inference. However, existing serving stacks typically commit to a static parallelism configuration at deployment; adapting to bursts, priorities, or long-context requests is often disruptive and slow. We present Flying Serving, a vLLM-based system that enables online DP-TP switching without restarting engine workers. Flying Serving makes reconfiguration practical by virtualizing the state that would otherwise force data movement: (i) a zero-copy Model Weights Manager that exposes TP shard views on demand, (ii) a KV Cache Adaptor that preserves request KV state across DP/TP layouts, (iii) an eagerly initialized Communicator Pool to amortize collective setup, and (iv) a deadlock-free scheduler that coordinates safe transitions under execution skew. Across three popular LLMs and realistic serving scenarios, Flying Serving improves performance by up to 4.79 × under high load and 3.47 × under low load while supporting latency- and memory-driven requests.

Gao, Shouwei [ORNL]↗
NASA NTRS1989

Bilinear modeling and nonlinear estimation

New methods are illustrated for online nonlinear estimation applied to the lateral deflection of an elastic beam on board measurements of angular rates and angular accelerations. The development of the filter equations, together with practical issues of their numerical solution as developed from global linearization by nonlinear output injection are contrasted with the usual method of the extended Kalman filter (EKF). It is shown how nonlinear estimation due to gyroscopic coupling can be implemented as an adaptive covariance filter using off-the-shelf Kalman filter algorithms. The effect of the global linearization by nonlinear output injection is to introduce a change of coordinates in which only the process noise covariance is to be updated in online implementation. This is in contrast to the computational approach which arises in EKF methods arising by local linearization with respect to the current conditional mean. Processing refinements for nonlinear estimation based on optimal, nonlinear interpolation between observations are also highlighted. In these methods the extrapolation of the process dynamics between measurement updates is obtained by replacing a transition matrix with an operator spline that is optimized off-line from responses to selected test inputs.

Dwyer, Thomas A. W., III↗
DOE OSTI2026

Microsecond-latency feedback at a particle accelerator by online reinforcement learning on hardware

The commissioning and operation of future large-scale scientific experiments will challenge current tuning and control methods. Reinforcement learning (RL) algorithms are a promising solution due to their ability to dynamically adapt to changing environments and consider delayed consequences. In many real-world applications, RL policies must produce actions in real time, often within microseconds to milliseconds, imposing significant constraints on system latency and computational overhead that conventional machine learning libraries are not designed to handle. To control phenomena in real time at these timescales, RL needs to be deployed on-the-edge, namely on dedicated hardware located near the system it controls, without relying on a host CPU or cloud-based inference. In this work we present the design and deployment of an experience accumulator system in a particle accelerator. In this system, deep-RL algorithms run using hardware acceleration and act within a few microseconds, enabling the use of RL for control of phenomena like beam instabilities. The training uses the collected data offline to reduce the number of operations carried out on the acceleration hardware. The proposed architecture was tested in real experimental conditions at the Karlsruhe research accelerator, a synchrotron light source, where the system was used to control artificially induced horizontal betatron oscillations in real-time, with a control loop period of just 2.7 μs. The results showed a performance comparable to the commercial feedback system available at the accelerator, demonstrating the viability and potential of this approach. Due to the self-learning and reconfiguration capability of this implementation, a seamless application to other control problems is possible. Applications range from particle accelerators to large-scale research and industrial facilities.

FPGA↗