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.

SourceResearch connectionFrom yearThrough year

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

At least 91 records · Page 5

All sources NASA NTRS DOE OSTI

DOE OSTI2024

The SciDAC QuantOm Framework: A composable Workflow

As part of the Scientific Discovery through Advanced Computing (SciDAC) program, the Quantum Chromodynamics Nuclear Tomography (QuantOM) project aims to analyze data from Deep Inelastic Scattering (DIS) experiments conducted at Jefferson Lab and the upcoming Electron Ion Collider. The DIS data analysis is performed on an event-level by leveraging nuclear theory models and accounting for experimental conditions. In order to efficiently run multiple analyses under varying conditions, a composable workflow was designed where each section (theory, experiment, objective minimization, etc.) has its own dedicated module. The optimization, i.e. the fit of theory to experimental data is carried out by deep learning techniques, such as Generative Adversarial Networks (GANs) or Reinforcement Learning (RL). This presentation gives an overview of the current status of the workflow, highlights present and future challenges, and highlights possible extensions to other projects with similar requirements.

Lersch, Daniel↗

DOE OSTI2024

Catalyst Layer Design, Manufacturing and In-line Quality Control

In this project we successfully demonstrated the capabilities of the Reactive Spray Deposition Technology (RSDT) to fabricate large-scale CCMs for advanced PEMWEs that have one-order of magnitude lower PGM loading in their catalyst layers, and performance comparable with the commercial state-of-the-art CCMs. The RSDT is a unique methodology that combines the catalyst synthesis and CCM fabrication in one step and reduces dramatically the time for CCM manufacturing. As fabricated large-scale CCMs with geometric area of 680 cm2 demonstrated excellent activity and durability performance, and the novel duo-recombination layer design paves the way for solving the safety concerns related to PEMWEs. In addition, excellent activity and durability performance has been demonstrated with RSDT fabricated CCMs with thinner membranes and duo RL design. This is a novel approach for further performance improvement of the MEAs for PEMWEs that has been successfully demonstrated for the first time in this project. The integration of the in-situ laser diagnostics system along with the in-line optical quality control system within the RSDT that has been achieved and demonstrated in this project, is an example for possibility of designing and building advanced manufacturing technologies that can meet the requirements of the future manufacturing. Therefore, the RSDT offers a precise real-time monitoring and control of the particles size, composition, loading, porosity, thickness, and defects in the catalysts’ layers, which render this technology as the best candidate for manufacturing of cost effective CCMs for PEMWEs. By using RSDT we successfully met all project’s milestones, Go/No-Go decision, objectives, goals, and deliverables.

08 HYDROGEN↗

DOE OSTI2025

Reinforcement Learning for In-Spill Optimization of the Mu2e Resonant Extraction: Compensating Non-Stationarity

We present design considerations and challenges for the fast machine learning component of a third-order resonant beam extraction regulation system being commissioned to deliver steady beam rates to the mu2e experiment at Fermilab. Dedicated quadrupoles drive the tune toward the 29/3 resonance each spill, extracting beam at kV multiwire septa. The overall Spill Regulation System consists of (1) a “slow” process using ~100-spill averages to adjust the base quad ramp infrequently, (2) a feedforward harmonic content compensator, and (3) the “fast” ML agent reacting during each ongoing spill with on-the-fly additive corrections to the sum of (1) and (2). We have demonstrated improved beam-rate steadying for a fast ML agent compared to a PID controller using a quasi-physical spill simulation, and demonstrated distillation of that simulation into a predictive surrogate model. Current work includes a data-and-training pipeline to generate data-aware surrogates with real-world dynamics, even as the dynamics shift unpredictably. The surrogates are to act as RL environments against which to train our fast ML control agents before deploying them on FPGA in the live system. Further current efforts focus on modeling and controlling beam loss around the storage ring, understanding additional available hardware inputs to the model, and the interplay of these with beam-steadying performance.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

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 OSTI2025

Replace Human Intelligence with Fast and Smart Geometric Reasoning and Graph Neural Network to Accelerate Next Gen ModSim Workflows

We present an agent-guided approach to CAD geometry decomposition that automates hex/hybrid meshing with graph neural networks (GNNs) to accelerate next-generation ModSim workflows. Our end-to-end pipeline (i) reduces 3D boundary-representation (B-Rep) models to a 2D chordal axis skeleton (CAT) and then to a 1D bipartite graph of surface and curve nodes, (ii) assigns per node labels as Cubit® WebCut actions, (iii) trains a multi-action GNN under supervised learning, and (iv) predicts five surface-node and three curve-node actions on out-of-distribution test geometries. Each graph node carries geometric, topological, and meshing attributes drawn from the B-Rep “skin” and CAT “skeleton,” with two-way mappings across 3D↔2D↔1D representations to maintain traceability back to 3D CAD. The supervised learning model exhibits stable convergence of the binary cross-entropy loss and achieves 98.7% accuracy on unseen lattice models. To operationalize decision-making, we rank predicted commands by geometric significance and prototyped the agent-guided workflow through the Cubit® Meshing PowerTool GUI. As a stretch goal, we explore reinforcement learning (RL) to reduce or remove label requirements and to learn policies for action sequences that maximize total reward (e.g., size of hex-meshable regions and resulting hex mesh quality). When all-hex meshing is not feasible, the agent assists in producing hybrid meshes—prioritizing hex in critical regions and transitioning to tetrahedral elements (tets) elsewhere—maintaining fidelity while ensuring robustness. The overarching objective is to replace manual, heuristics-based decomposition with data-driven, reproducible automation, cutting meshing turnaround time by orders of magnitude. We anticipate direct impact on simulation workflows through intelligent, scalable decomposition of complex CAD models into hex-meshable subdomains.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Closing the Loop between In Situ Stress Complexity and EGS Fracture Complexity

42 ENGINEERING↗

DOE OSTI2025

REBOUND: Reverse Engineering Bidirectional Outflow Under Non-Equilibrium Diffusion

Rare-earth elements (REEs) are essential for electronics, renewable energy, and defense technologies. However, the current supply of REEs relies on mining concentrated in a few countries and energy-intensive separations. DOE’s Basic Energy Sciences (BES) program has launched a grand challenge which aims to ensure a sustainable supply of critical REEs by developing innovative and environmentally friendly separation methods. As an alternative to costly and harmful traditional methods, the Non-Equilibrium Transport Driven Separations (NETS) initiative has created a microfluidic Y-channel co-flow method that applies external fields to exploit magneto- and electrohydrodynamic effects for separating dilute REE ions from complex feedstocks. Computational fluid dynamics (CFD) studies have identified a few operating conditions with promising ion selectivity and separation efficiency. However, challenges remain regarding Y-channel versatility across feedstocks and accurate incorporation of physical phenomena into CFD models. In this work, we develop a multi-fidelity modelling approach which integrates experimental results with CFD simulation to build a surrogate model for the dependence of separation efficiency to variation of design parameters. The surrogate model enables a reinforcement learning (RL) method to adaptively launch CFD and experimental runs, improving model fidelity around optimal Y-channel parameters.

36 MATERIALS SCIENCE↗

DOE OSTI2024

Summer Bat Monitoring Report for Calendar Year 2023

Bats are mammals belonging to the order Chiroptera, which means “hand-wing.” They use a thin membrane of skin stretched between their fingers to fly. All bat species known in the state of Washington are insectivorous and each bat can consume 600-1,000 insects per hour while feeding. Bats typically give birth once per year to a single pup, making bat populations vulnerable to impacts and slow to recover. Several species of bats have been documented on the Hanford Site, with nine species identified during The Nature Conservancy surveys in 1997 and 1998, and an additional seven species listed as potentially present as documented in Biodiversity Inventory and Analysis of the Hanford Site. Final Report 1994-1999 (Soll et al., 1999). Ongoing U.S. Department of Energy (DOE), Richland Operations Office (RL) acoustic surveys, including HNF-53759, Summer Bat Monitoring Report for Calendar Year 2012, have also confirmed the presence of the same nine species documented by Soll et al. (1999). An additional species, the spotted bat (Euderma maculatum), was observed roosting on the exterior of a facility by Hanford Site personnel in 2021; the species was confirmed by Hanford Site biologists via photograph (Table 1-1).

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Employing MACS/ViBRANT as a Surrogate MARVEL Reactor for Startup Reactivity Tuning and Supervisory Control Processes

Advanced nuclear reactors are a key part of the future of nuclear energy both in the United States and globally. They offer unique benefits for various energy-demanding applications, including use in remote locations, compact size, modular manufacturing, remote monitoring, low and/or variable power rating operation, and reliance on novel technologies to enhance operational safety. To achieve economic feasibility, advanced reactors must significantly reduce their workforces in comparison with the current fleet. Achieving this reduction will occur through reducing staff workloads using technology to achieve autonomous or semi-autonomous operations, demonstrated by comprehensive testing and validation activities. These operations will require both software and hardware platforms during the design and testing phases. While simulations are useful during the design phase, their performance can significantly deviate during actual deployment on hardware. This report presents the outcomes of a collaborative technical initiative between the U.S. Department of Energy (DOE) Microreactor Program (MRP) and Advanced Sensors and Instrumentation (ASI) Program. The collaboration utilized the Microreactor Automated Control System (MACS) hardware platform to bridge the gap between theoretical reactor design and actual startup and control operations. Two key use cases were investigated: facilitating the startup testing period and demonstrating supervisory control. The first use case details the key Microreactor Applications Research Validation and Evaluation (MARVEL) reactor startup physics testing activities conducted using the MACS platform. These activities included drum worth measurements, shutdown margin assessment, temperature feedback analysis, and scram time evaluation, as well as unique testing that would apply to the MARVEL reactor to demonstrate the testing methodologies in a low-risk environment. The MACS platform, serving as a surrogate representation of the MARVEL reactor, proved instrumental in performing these tests. The exercise revealed aspects that led to optimized processes, refined hardware design, and enhanced base software capabilities. By maturing methods and technologies in this manner, the initiative promises to reduce wasted time in the actual on-site reactor deployment effort, thereby saving significant time and resources. The second use case focuses on the development and implementation of supervisory control methods aimed at managing core tilt, which can result from asymmetrical operations or manufacturing imperfections in fuel rods or reactivity control devices. A key objective was to assess and compare the use of artificial intelligence (AI) for supervisory control. The effort aimed to define the role of supervisory control to enhance performance without risking control instability. This effort explored three distinct approaches: rules-based (RB) methods, optimization techniques, and reinforcement learning (RL) algorithms. Each approach was evaluated for its ease of implementation, its usability, and its effectiveness in responding to asymmetries in neutron flux. Comparative analysis of these approaches provided valuable insights into their applicability and effectiveness, offering a robust framework for advanced reactor operations. Together, these two use cases highlight the potential of hardware test beds to help streamline the design, operation, and control of advanced nuclear reactors. This collaborative effort underscores the importance of continued innovation and experimentation in achieving the next generation of safe, reliable, and economically viable nuclear energy solutions.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2026

Planetary Boundary-Layer Height (PBLHT) Value-Added Product: Remote-Sensing Retrievals

The planetary boundary layer (PBL) is fundamental to numerous atmospheric processes, including aerosol mixing and transport, cloud evolution, and precipitation formation. A critical parameter in these studies is the PBL height (PBLHT). This vertical depth is essential for characterizing PBL structures in numerical simulations and serves as a primary metric for estimating flux exchanges between the Earth’s surface and the atmosphere. Radiosonde (SONDE) observations provide high-vertical-resolution measurements of temperature and moisture profiles and are widely used to estimate PBLHT (Liu and Liang 2010, Seidel et al. 2010). The U.S. Department of Energy Atmospheric Radiation Measurement (ARM) User Facility’s PBLHT value-added product (VAP) for radiosonde measurements, known as PBLHTSONDE, applies three commonly used methods—the Heffter (1980) method, the Liu and Liang (2010) method, and the bulk Richardson number approach (Seibert et al. 2000)—to derive PBLHT. The PBLHTSONDE VAP operates routinely at ARM observatories and mobile facilities, with data available from the ARM Data Center shortly after sounding observations are collected (Sivaraman et al. 2013). However, radiosonde observations are limited by their low temporal resolution. Most stations launch soundings only twice daily, which constrains the ability to investigate and characterize the temporal evolution of the PBL using radiosonde data alone. The use of continuous remote-sensing observations provides high temporal resolution of PBLHT estimates. These observations include aerosol lidars (Dang et al. 2019, Su et al. 2020), Doppler lidar (DL; Tucker et al. 2009, Krishnamurthy et al. 2021), and water vapor and/or temperature lidars and radiometers (Turner et al. 2014). These observations provide valuable data on the PBL’s thermodynamic properties (e.g., water vapor and/or temperature lidars and radiometers), dynamic properties (e.g., DL), and distribution of tracer substances (e.g., aerosol lidars), all of which can be used to estimate PBLHT. ARM developed PBLHT estimates from the micropulse lidar (MPL; PBLHTMPL), Doppler lidar (PBLHTDL), and combined Raman lidar (RL)/atmospheric emitted radiance interferometer (AERI) thermodynamic profiles (PBLHTTHERMO). Each estimate captures different physical characteristics of the boundary layer—aerosol tracers, vertical velocity turbulence, and thermodynamic structure—and exhibits distinct strengths and limitations depending on the PBL regime and time of day. In addition, the ARM ceilometer (CEIL) provides three potential PBLHT candidates derived from the vendor's built-in algorithm. Building on these individual retrievals, ARM developed the PBLHTBEML VAP, which combines the four remote-sensing-based estimates with ancillary meteorological variables using the machine learning approach of Zhang et al. (2025) to produce a best-estimate PBLHT at 10-minute resolution.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2024

Reinforcement Learning Control for Enhancing Marine Hydrokinetic Turbine Energy Generation

This paper proposes a reinforcement learning-based method to maximize power generation for a direct-drive marine hydrokinetic turbine. A high levelized cost of energy (LCOE) is preventative in the widespread adoption of many marine energy conversion technologies. A straightforward way to reduce LCOE is to increase conversion efficiency and ensure maximum energy generation. The proposed method utilizes a damping control methodology, varying applied generator torque via a linear relationship between the applied damping coefficient and rotor speed. A state-action-reward-state-action (SARSA) algorithm has been used to learn the optimal control action for a given flow velocity. The proposed SARSA methodology uses Gaussian radial basis functions to create a three-dimensional surface to estimate the relationship between damping coefficient, incoming flow velocity, and coefficient of power (C p ). Here, the SARSA algorithm was compared against a baseline optimal tip speed ratio controller over a year-long flow velocity case profile while considering the effects of biofouling on the turbine system, where the proposed RL method generated 0.92% more energy than the baseline.

Damp↗

DOE OSTI2024

KBase Narrative - Formation of a constructed microbial community in a nutrient rich environment indicates bacterial interspecific competition

This Narrative is the parent for the three Narratives referenced in "Formation of a constructed microbial community in a nutrient rich environment indicates bacterial interspecific competition." Wang J, Appidi MR, Burdick LH, Abraham PE, Hettich RL, Pelletier DA, Doktycz MJ. 2024. Formation of a constructed microbial community in a nutrient-rich environment indicates bacterial interspecific competition. mSystems. American Society for Microbiology.

Wang, Jia↗

DOE OSTI2025

Unconventional compute methods and future challenges for superconducting digital computing

Superconducting digital computing (SDC) based on Josephson junctions (JJs) offers significant potential for enhancing compute throughput and reducing energy consumption compared to conventional room-temperature CMOS-based approaches. Current superconducting logic families exhibit diverse characteristics in clocking strategies, power management, and information encoding techniques. This paper reviews recent advancements in unconventional computing methods specifically designed for superconducting digital circuits, emphasizing temporal computing and pulse-train representations. Notable techniques include race logic (RL), temporal pulse train computing (U-SFQ), and temporal multipliers, each offering unique performance and area advantages suited to superconducting implementations. Additionally, this paper reviews innovations in superconducting coarse-grain reconfigurable architectures (CGRA), superconducting-specific on-chip communication architectures, cryogenic sensor interfaces, and quantum computing control electronics. Finally, we highlight research challenges that should be addressed to facilitate the widespread adoption of superconducting digital computing.

EDA tools↗

DOE OSTI2025

ARM-IRL: Adaptive Resilience Metric Quantification Using Inverse Reinforcement Learning

The resilience of safety-critical systems is gaining importance due to the rise in cyber and physical threats, especially within critical infrastructure. Traditional static resilience metrics may not capture dynamic system states, leading to inaccurate assessments and ineffective responses to cyber threats. This work aims to develop a data-driven, adaptive method for resilience metric learning. We propose a data-driven approach using inverse reinforcement learning (IRL) to learn a single, adaptive resilience metric. The method infers a reward function from expert control actions. Unlike previous approaches using static weights or fuzzy logic, this work applies adversarial inverse reinforcement learning (AIRL), training a generator and discriminator in parallel to learn the reward structure and derive an optimal policy. The proposed approach is evaluated on multiple scenarios: optimal communication network rerouting, power distribution network reconfiguration, and cyber–physical restoration of critical loads using the IEEE 123-bus system. The adaptive, learned resilience metric enables faster critical load restoration in comparison to conventional RL approaches.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Integration of Decentralized Graph-Based Multi-Agent Reinforcement Learning with Digital Twin for Traffic Signal Optimization

Machine learning (ML) methods, particularly Reinforcement Learning (RL), have gained widespread attention for optimizing traffic signal control in intelligent transportation systems. However, existing ML approaches often exhibit limitations in scalability and adaptability, particularly within large traffic networks. This paper introduces an innovative solution by integrating decentralized graph-based multi-agent reinforcement learning (DGMARL) with a Digital Twin to enhance traffic signal optimization, targeting the reduction of traffic congestion and network-wide fuel consumption associated with vehicle stops and stop delays. In this approach, DGMARL agents are employed to learn traffic state patterns and make informed decisions regarding traffic signal control. The integration with a Digital Twin module further facilitates this process by simulating and replicating the real-time asymmetric traffic behaviors of a complex traffic network. The evaluation of this proposed methodology utilized PTV-Vissim, a traffic simulation software, which also serves as the simulation engine for the Digital Twin. The study focused on the Martin Luther King (MLK) Smart Corridor in Chattanooga, Tennessee, USA, by considering symmetric and asymmetric road layouts and traffic conditions. Comparative analysis against an actuated signal control baseline approach revealed significant improvements. Experiment results demonstrate a remarkable 55.38% reduction in Eco_PI, a developed performance measure capturing the cumulative impact of stops and penalized stop delays on fuel consumption, over a 24 h scenario. In a PM-peak-hour scenario, the average reduction in Eco_PI reached 38.94%, indicating the substantial improvement achieved in optimizing traffic flow and reducing fuel consumption during high-demand periods. These findings underscore the effectiveness of the integrated DGMARL and Digital Twin approach in optimizing traffic signals, contributing to a more sustainable and efficient traffic management system.

42 ENGINEERING↗

DOE OSTI2025

GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning

Large language models (LLMs) are increasingly adapted to downstream tasks via reinforcement learning (RL) methods like Group Relative Policy Optimization (GRPO), which often require thousands of rollouts to learn new tasks. We argue that the interpretable nature of language often provides a much richer learning medium for LLMs, compared to policy gradients derived from sparse, scalar rewards. To test this, we introduce GEPA (Genetic-Pareto), a prompt optimizer that thoroughly incorporates natural language reflection to learn high-level rules from trial and error. Given any AI system containing one or more LLM prompts, GEPA samples trajectories (e.g., reasoning, tool calls, and tool outputs) and reflects on them in natural language to diagnose problems, propose and test prompt updates, and combine complementary lessons from the Pareto frontier of its own attempts. As a result of GEPA's design, it can often turn even just a few rollouts into a large quality gain. Across six tasks, GEPA outperforms GRPO by 6% on average and by up to 20%, while using up to 35x fewer rollouts. GEPA also outperforms the leading prompt optimizer, MIPROv2, by over 10% (e.g., +12% accuracy on AIME-2025), and demonstrates promising results as an inference-time search strategy for code optimization. We release our code at https://github.com/gepa-ai/gepa.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Best estimate of the planetary boundary layer height from multiple remote sensing measurements

Remote sensing measurements have been widely used to estimate the planetary boundary layer height (PBLHT). Each remote sensing approach offers unique strengths and faces different limitations. In this study, we use machine learning (ML) methods to produce a best-estimate PBLHT (PBLHT-BE-ML) by integrating four PBLHT estimates derived from remote sensing measurements at the Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) Southern Great Plains (SGP) observatory. Three ML models – random forest (RF) classifier, RF regressor, and light gradient-boosting machine (LightGBM) – were trained on a dataset from 2017 to 2023 that included radiosonde, various remote sensing PBLHT estimates, and atmospheric meteorological conditions. Evaluations indicated that PBLHT-BE-ML from all three models improved alignment with the PBLHT derived from radiosonde data (PBLHT-SONDE), with LightGBM demonstrating the highest accuracy under both stable and unstable boundary layer conditions. Feature analysis revealed that the most influential input features at the SGP site were the PBLHT estimates derived from (a) potential temperature profiles retrieved using Raman lidar (RL) and atmospheric emitted radiance interferometer (AERI) measurements (PBLHT-THERMO), (b) vertical velocity variance profiles from Doppler lidar (PBLHT-DL), and (c) aerosol backscatter profiles from micropulse lidar (PBLHT-MPL). The trained models were then used to predict PBLHT-BE-ML at a temporal resolution of 10 min, effectively capturing the diurnal evolution of PBLHT and its significant seasonal variations, with the largest diurnal variation observed over summer at the SGP site. We applied these trained models to data from the ARM Eastern Pacific Cloud Aerosol Precipitation Experiment (EPCAPE) field campaign (EPC), where the PBLHT-BE-ML, particularly with the LightGBM model, demonstrated improved accuracy against PBLHT-SONDE. Analyses of model performance at both the SGP and EPC sites suggest that expanding the training dataset to include various surface types, such as ocean and ice-covered areas, could further enhance ML model performance for PBLHT estimation across varied geographic regions.

Zhang, Damao [Pacific Northwest National Laborator↗

DOE OSTI2024

"Traffic Control via Connected and Automated Vehicles: An Open-Road Field Experiment with 100 CAVs"

The CIRCLES project aims to reduce instabilities in traffic flow, which are naturally occurring phenomena due to human driving behavior. These "phantom jams" or "stop-and-go waves,"are a significant source of wasted energy. Toward this goal, the CIRCLES project designed a control system referred to as the MegaController by the CIRCLES team, that could be deployed in real traffic. Our field experiment leveraged a heterogeneous fleet of 100 longitudinally-controlled vehicles as Lagrangian traffic actuators, each of which ran a controller with the architecture described in this paper. The MegaController is a hierarchical control architecture, which consists of two main layers. The upper layer is called Speed Planner, and is a centralized optimal control algorithm. It assigns speed targets to the vehicles, conveyed through the LTE cellular network. The lower layer is a control layer, running on each vehicle. It performs local actuation by overriding the stock adaptive cruise controller, using the stock on-board sensors. The Speed Planner ingests live data feeds provided by third parties, as well as data from our own control vehicles, and uses both to perform the speed assignment. The architecture of the speed planner allows for modular use of standard control techniques, such as optimal control, model predictive control, kernel methods and others, including Deep RL, model predictive control and explicit controllers. Depending on the vehicle architecture, all onboard sensing data can be accessed by the local controllers, or only some. Control inputs vary across different automakers, with inputs ranging from torque or acceleration requests for some cars, and electronic selection of ACC set points in others. The proposed architecture allows for the combination of all possible settings proposed above. Most configurations were tested throughout the ramp up to the MegaVandertest.

Lee, Jonathan↗