Search NASA⌕ Search

SEARCH · Search NASA

Results for “practical reinforcement learning”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Revisiting the ODE Method for Recursive Algorithms: Fast Convergence Using Quasi Stochastic Approximation

Several decades ago, Profs. Sean Meyn and Lei Guo were postdoctoral fellows at ANU, where they shared interest in recursive algorithms. It seems fitting to celebrate Lei Guo's 60th birthday with a review of the ODE Method and its recent evolution. The method has been regarded as a technique for algorithm analysis. It is argued that this viewpoint is backwards: The original stochastic approximation method was surely motivated by an ODE, and tools for analysis came much later (based on establishing robustness of Euler approximations). The paper presents a brief survey of recent research in machine learning that shows the power of algorithm design in continuous time, following by careful approximation to obtain a practical recursive algorithm. While these methods are usually presented in a stochastic setting, this is not a prerequisite. In fact, recent theory shows that rates of convergence can be dramatically accelerated by applying techniques inspired by quasi Monte-Carlo. Subject to conditions, the optimal rate of convergence can be obtained by applying the averaging technique of Polyak and Ruppert. The conditions are not universal, but theory suggests alternatives to achieve acceleration. The theory is illustrated with applications to gradient-free optimization, and policy gradient algorithms for reinforcement learning.

learning and adaptive systems in artificial intell↗
DOE OSTI2022

Network Reconfiguration for Enhanced Operational Resilience Using Reinforcement Learning

This paper proposes a reinforcement learning-based approach for distribution network reconfiguration(DNR) to enhance the resilience of the electric power supply. Resilience enhancements usually require solving large-scale stochastic optimization problems that are computationally expensive and sometimes infeasible. The exceptional performance of reinforcement learning techniques has encouraged their adoption in various power system control studies, specifically resilience-based real-time applications. In this paper, a single agent framework is developed using an Actor-Critic algorithm (ACA) to determine statuses of tie-switches in a distribution feeder impacted by an extreme weather event. The proposed approach provides a fast-acting control algorithm that reconfigures the feeder topology to reduce or even avoid load shedding. The problem is formulated as a discrete Markov decision process in such a way that a system state captures the system topology and its operational characteristics. An action is made to open or close a specific set of tie-switches after which a reward is calculated to evaluate the practicality and advantage of that action. The iterative Markov process is used to train the proposed ACA under diverse failure scenarios and is demonstrated on the 33-node distribution feeder system. Results show the capability of the proposed ACA to determine proper switching action of tie-switches with accuracy exceeding 93%.

actor critic↗
DOE OSTI2023

Efficient learning of power grid voltage control strategies via model-based deep reinforcement learning

Here this article proposes a model-based deep reinforcement learning (DRL) method to design emergency control strategies for short-term voltage stability problems in power systems. Recent advances show promising results for model-free DRL-based methods in power systems control problems. But in power systems applications, these model-free methods have certain issues related to training time (clock time) and sample efficiency; both are critical for making state-of-the-art DRL algorithms practically applicable. DRL-agent learns an optimal policy via a trial-and-error method while interacting with the real-world environment. It is also desirable to minimize the direct interaction of the DRL agent with the real-world power grid due to its safety-critical nature. Additionally, the state-of-the-art DRL-based policies are mostly trained using a physics-based grid simulator where dynamic simulation is computationally intensive, lowering the training efficiency. We propose a novel model-based DRL framework where a deep neural network (DNN)-based dynamic surrogate model (SM), instead of a real-world power grid or physics-based simulation, is utilized within the policy learning framework, making the process faster and more sample efficient. However, having stable training in model-based DRL is challenging because of the complex system dynamics of large-scale power systems. We addressed these issues by incorporating imitation learning to have a warm start in policy learning, reward-shaping, and multi-step loss in surrogate model training. Finally, we achieved 97.5% reduction in samples and 87.7% reduction in training time for an application to the IEEE 300-bus test system.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

On the Verification of Deep Reinforcement Learning Solution for Intelligent Operation of Distribution Grids

Capabilities of deep reinforcement learning (DRL) in obtaining fast decision policies in high dimensional and stochastic environments have led to its extensive use in operational research, including the operation of distribution grids with high penetration of distributed energy resources (DER). However, the feasibility and robustness of DRL solutions are not guaranteed for the system operator, and hence, those solutions may be of limited practical value. This paper proposes an analytical method to find feasibility ellipsoids that represent the range of multi-dimensional system states in which the DRL solution is guaranteed to be feasible. Empirical studies and stochastic sampling determine the ratio of the discovered to the actual feasible space as a function of the sample size. In addition, the performance of logarithmic, linear, and exponential penalization of infeasibility during the DRL training are studied and compared in order to reduce the number of infeasible solutions

Hosseini, Mohammad Mehdi↗
DOE OSTI2024

Learning Constrained Parametric Differentiable Predictive Control Policies With Guarantees

We present differentiable predictive control (DPC), a method for offline learning of constrained neural control policies for nonlinear dynamical systems with performance guarantees. We show that the sensitivities of the parametric optimal control problem can be used to obtain direct policy gradients. Specifically, we employ automatic differentiation (AD) to efficiently compute the sensitivities of the model predictive control (MPC) objective function and constraints penalties. To guarantee safety upon deployment, we derive probabilistic guarantees on closed-loop stability and constraint satisfaction based on indicator functions and Hoeffding’s inequality. We empirically demonstrate that the proposed method can learn neural control policies for various parametric optimal control tasks. In particular, we show that the proposed DPC method can stabilize systems with unstable dynamics, track time-varying references, and satisfy nonlinear state and input constraints. Our DPC method has practical time savings compared to alternative approaches for fast and memory-efficient controller design. Specifically, DPC does not depend on a supervisory controller as opposed to approximate MPC based on imitation learning. We demonstrate that, without losing performance, DPC is scalable with greatly reduced demands on memory and computation compared to implicit and explicit MPC while being more sample efficient than model-free reinforcement learning (RL) algorithms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Physics-Informed Neural Network (PINN) Prediction of Mixed Mass-Heat-Crystallization Limited Methane Hydrate Formation and Dissociation in Micro-Confinement

The creation and use of Physics-Informed Neural Networks (PINNs) for simulating the dynamics of methane hydrate formation and dissociation will be presented. The PINN framework's main benefit is its capacity to impose physical consistency with only a partial comprehension of the governing equations. This makes the algorithm especially useful for systems with little experimental evidence or a lack of theoretical knowledge. A strong basis for forecasting methane hydrate behavior over the verified operating ranges of 30.0-80.9 bar pressure and 1.0-4.0 K sub-cooling conditions is provided by the combination of conductive heat transfer equations and mixed mass-transfer–crystallization kinetics. PINNs were more accurate at predicting the mixed mass-heat-crystallization limited kinetics than conventional Artificial Neural Networks (ANNs), demonstrating remarkable predictive accuracy for methane hydrate production over the ANN model. The efficiency of incorporating physical limitations from first principles into machine learning frameworks for methane hydrate crystallizations is reinforced by these findings. For hydrate-related applications in energy generation, carbon sequestration, and climate modelling, our study establishes PINNs as a computational tool that is both scalable and efficient. The proven capacity to close the gap between conventional physics-based simulations and solely data-driven models creates new opportunities for expedited hydrate research and practical applications.

Hartman, Ryan L [NYU Tandon School of Engineering]↗
DOE OSTI2025

A Sequential Model Predictive and Deep Reinforcement Learning-Based Controller for Distribution System Outage Mitigation under Hurricane Events

This paper proposes a proactive outage mitigation framework for power distribution networks to withstand hurricane-induced disruptions. It leverages Model Predictive Control (MPC) to identify safe lines for proactive switching during hurricanes, minimizing the risk of cascading failures and voltage violations. The switching strategies optimized by MPC are sequentially integrated with a Deep Reinforcement Learning agent using the Advantage Actor-Critic algorithm, enabling dynamic line switching to maximize connected buses and minimize voltage violations in real time. Using a probabilistic hurricane model, the framework predicts line failures and adapts to varying conditions to enhance grid resilience. Simulations on the IEEE 123-bus system demonstrate its effectiveness in maintaining high connectivity and minimizing disruptions. Real-time testing with an RTDS confirms the practicality and reliability of the proposed approach.

Selim, Alaa [University of Connecticut]↗
DOE OSTI2022

AI-Enabled Discovery and Physics-Based Optimization of Energy Efficient Processing Strategies for Advanced Turbine Alloys (Final Technical Report)

In this project, the multi-organizational team of academic and industrial researchers from the University of Kentucky an aerospace and energy generation OEM partner has leveraged novel Digital Process Twin (DPT) models of process/structure interactions (i.e., process-induced surface integrity) to advance a paradigm of fully integrated computational materials engineering (ICME). Using efficient process models as the core of a digital process simulator for a reinforcement learning algorithm, the team has integrated industrial data and metrics of structure/performance/energy relationships and manufacturing-related energy metrics to optimize dynamic processing parameters for significantly improved life-cycle energy efficiency of advanced γ-TiAl low-pressure turbine (LPT) alloys, as indicated by a set of design relevant parameters (e.g., residual stresses and scrap rate). The key objective and anticipated outcome of the project was at least a 10% reduction in life-cycle embodied energy for a recently developed, γ-TiAl low-pressure turbine (LPT) alloy and nickel-based superalloy Inconel 718, through the adoption of the proposed AI-enabled process optimization approach. The final project outcomes significantly exceeded this original target, realizing manufacturing-related energy efficiency improvements of more than 130% for TiAl and up to 80% for Inconel 718. Rather than following the prevailing and highly inefficient empirical paradigm, the proposed study demonstrated the feasibility of adopting a digital, physics-based process design and optimization paradigm. The recurring need for manual intervention, rework, reinspection causes significant production bottlenecks and unnecessary expense associated with delivering the requisite component quality. The OEM partner, and turbine industry in general, expect to reap significant cost and resource savings if an AI-optimized set of parameters can be applied to specific machining operations. The technical scope of the proposed project involved the paving of a realistic path towards model-based and AI-enabled Integrated Computational Materials Engineering (ICME), and away from inefficient empirical process optimization and legacy manufacturing practices, which are no longer able to efficiently process novel high-performance turbine alloy materials. The project team will address the fundamental knowledge gap that currently exists within the ICME paradigm with respect to the process/structure/performance/energy impacts of finishing processes. While significant resources have been devoted to the ‘early stages’ of manufacturing, such as alloy design, primary and secondary processing, finishing processes have not been adequately integrated within ICME. To provide an actionable path towards model-based finishing process design (e.g., machining, burnishing, grinding, polishing), we will employ a novel AI-enabled process optimization paradigm, based on a computationally efficient, physics-based process simulator. Through limited experimental work to calibrate and validate our process simulator model via an advanced in-situ characterization technique and process optimization via reinforcement learning, the project will seek to demonstrate a viable alternative to the inefficient ‘legacy’ processing strategies, empirical testing and broad scope machining learning approaches, all of which fail to adequately consider complex process physics. The project team has identified an intermetallic γ-TiAl LPT alloy, which is currently being used as part of the OEM partner’s advanced gas turbine designs. This particular alloy poses significant manufacturing challenges during finishing operations, which limit the degree to which the current turbine design can be manufactured in an energy- and cost-efficient manner. Empirical testing and numerical modeling efforts to optimize processing parameters for γ-TiAl have not been able to resolve these manufacturing challenges, so the proposed physics-based AI-enabled optimization technology would offer a truly novel and transformative capability. The multi-organizational team of academic and industry experts from the UKY and the OEM partner will work together closely to demonstrate the analytical and experimental critical function and characteristic proof of concept of this novel approach.

20 FOSSIL-FUELED POWER PLANTS↗
DOE OSTI2022

A resilient network recovery framework against cascading failures with deep graph learning

Because of the increasing importance and dependencies of infrastructure networks and the potential for massive cascading failures in real-world network systems, maintenance optimization to effectively reduce system performance loss caused by diverse disruptions is of significant interest among researchers and practitioners. In this work, a new recovery framework was developed to rapidly identify important system components for maintenance to improve network resilience against cascading failures. Here this work provides distinct advantages to determine an optimal maintenance priority by combining real-time network structure importance with other maintenance prioritization based on customer preference. This approach adopts structural graph embedding and deep reinforcement learning to extract real-time network topology information (such as minimum vertex cover) to update the maintenance priority during the recovery process. Based on the case studies on synthetic networks and a US airport network, the proposed recovery framework with real-time network topology awareness shows better performance than other maintenance prioritization strategies regarding resilience enhancement. This work improves the understanding of how the changing network structure influences maintenance effects. It also provides insights of the practical usefulness of advanced deep learning on helping optimal maintenance prioritization to effectively reduce the intensity and extent of cascading failures.

42 ENGINEERING↗
DOE OSTI2022

A reinforcement learning approach to long-horizon operations, health, and maintenance supervisory control of advanced energy systems

In this work, we develop a Reinforcement Learning (RL) approach to the supervisory control problem for advanced energy systems, such as novel nuclear reactors and other demand-driven, mission-critical, and component-health-sensitive energy plants. The inclusive problem landscape considered captures the stochastic confluence of plant performance, component health evolution, power demand from the grid, diverse maintenance actions, and operator-defined goals and constraints, all considered over meaningfully long-enough reasoning horizons. Key aspects of the proposed approach are a receding horizon control-inspired technique dictating time- or event-triggered supervisory policy (re-)constructions, as well as additional capability-enabling contributions such as timescale compression, to handle long reasoning horizons and uncertainty in parts of the problem, and practical yet demonstrably-effective handling of hybrid action spaces with continuous and discrete decision variables. The resulting algorithm consists of a simulation-based RL agent constructing stochastic supervisory control policies over nontrivial action spaces and for long horizons, applying the learned policy to the system for a much shorter interval, and perpetually repeating, to construct the next long-horizon policy. That next policy will only be applied, again, for a short interval, yet originally far-in-time events move progressively closer, their associated uncertainty decreases, and new events and aspects enter the reasoning horizon. The proposed methodology bridges fundamental receding horizon concepts with the unequivocally stronger and more scalable reasoning of contemporary RL. Numerical examples using Soft Actor–Critic Deep RL illustrate the operation and efficacy of the proposed technique for a power plant tasked with health-aware load following missions in a dynamic electricity market landscape.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Adaptive nonequilibrium design of actin-based metamaterials: Fundamental and practical limits of control

The adaptive and surprising emergent properties of biological materials self-assembled in far-from-equilibrium environments serve as an inspiration for efforts to design nanomaterials. In particular, controlling the conditions of self-assembly can modulate material properties, but there is no systematic understanding of either how to parameterize external control or how controllable a given material can be. Here, we demonstrate that branched actin networks can be encoded with metamaterial properties by dynamically controlling the applied force under which they grow and that the protocols can be selected using multi-task reinforcement learning. These actin networks have tunable responses over a large dynamic range depending on the chosen external protocol, providing a pathway to encoding “memory” within these structures. Interestingly, we obtain a bound that relates the dissipation rate and the rate of “encoding” that gives insight into the constraints on control—both physical and information theoretical. Taken together, these results emphasize the utility and necessity of nonequilibrium control for designing self-assembled nanostructures.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

FY20 Proxy App Suite Release: Report for ECP Proxy App Project Milestone ADCD-504-10

Version 4.0 of the ECP Proxy App Suite is practically unchanged from the previous release. The current set of proxies has proven useful for many aspects of benchmarking and co-design and we see little reason to alter the suite. Although there have been few changes to the ECP suite, the team has been hard at work in other areas. In the area of Machine Learning (ML) we have now created a separate proxy suite dedicated to this scientific applications of ML. The suite includes: miniGAN (Generative Adversarial Networks), miniRL (Reinforcement Learning), CRADL (inline inference), Cosmoflow-Benchmark (Convolutional Neural Network), and MLPerf-DeepCam (Climate Segmentation Benchmark). Section 3 contains more information about these proxies as well as the principles that are guiding the development of the suite. We have surveyed available proxies for several application domains including Computational Fluid Dynamics, Quantum Chemistry, Quantum Computing Simulation, Molecular Dynamics, Monte Carlo Transport, and Density Functional Theory to identify gaps in proxy coverage. Several new proxy apps are either already available or will be released soon to fill these gaps. Section 4 provides full details. Finally, section 5 reports on our continued collaboration with the ECP Continuous Integration (CI) effort to use proxy apps to help identify problems and roadblocks to cross-lab CI. We have also assisted the El Capitan Center of Excellence (COE) to stand up a CI system that will be used to test releases of HPE and AMD software stacks. We hope that the COE effort can serve as a model for ECP by showing how the Proxy App Team can work with AD and ST teams identify critical features, kernels, patterns, etc. and incorporate them into a CI system that will help ensure that Frontier and Aurora will provide those needed capabilities.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Real-time quantum error correction beyond break-even

The ambition of harnessing the quantum for computation is at odds with the fundamental phenomenon of decoherence. The purpose of quantum error correction (QEC) is to counteract the natural tendency of a complex system to decohere. This cooperative process, which requires participation of multiple quantum and classical components, creates a special type of dissipation that removes the entropy caused by the errors faster than the rate at which these errors corrupt the stored quantum information. Previous experimental attempts to engineer such a process faced an excessive generation of errors that overwhelmed the error-correcting capability of the process itself. Whether it is practically possible to utilize QEC for extending quantum coherence thus remains an open question. We answer it by demonstrating a fully stabilized and error-corrected logical qubit whose quantum coherence is significantly longer than that of all the imperfect quantum components involved in the QEC process, beating the best of them with a coherence gain of G=2.27±0.07. Here we achieve this performance by combining innovations in several domains including the fabrication of superconducting quantum circuits and model-free reinforcement learning.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

RLScheduler: An Automated HPC Batch Job Scheduler Using Reinforcement Learning

Today’s high-performance computing (HPC) platforms are still dominated by batch jobs. Accordingly, effective batch job scheduling is crucial to obtain high system efficiency. Existing HPC batch job schedulers typically leverage heuristic priority functions to prioritize and schedule jobs. But, once configured and deployed by the experts, such priority functions can hardly adapt to the changes of job loads, optimization goals, or system settings, potentially leading to degraded system efficiency when changes occur. To address this fundamental issue, we present RLScheduler, an automated HPC batch job scheduler built on reinforcement learning. RLScheduler relies on minimal manual interventions or expert knowledge, but can learn high-quality scheduling policies via its own continuous ‘trial and error’. We introduce a new kernel-based neural network structure and trajectory filtering mechanism in RLScheduler to improve and stabilize the learning process. Through extensive evaluations, we confirm that RLScheduler can learn high-quality scheduling policies towards various workloads and various optimization goals with relatively low computation cost. Moreover, we show that the learned models perform stably even when applied to unseen workloads, making them practical for production use.

Zhang, Di↗
DOE OSTI2025

A Two-Stage Quantum Reinforcement Learning Method for Multi-Objective Transmission Switching

Multi-objective transmission switching (MO-TS) problems involve the strategic reconfiguration of network topology to simultaneously optimize multiple objectives. As the system scale increases, finding feasible solutions becomes increasingly challenging due to the problem's nonlinearity and high computational complexity. To address these challenges, this paper proposes a two-stage quantum reinforcement learning method that leverages potential quantum advantages for MO-TS. In the first stage, candidate switching lines are identified using a graph-theoretical approach to reduce the problem's dimensionality. The second stage introduces a quantum-classical reinforcement learning framework, where a learnable measurement-based CNN-ResVQC architecture is developed to effectively reduce the input dimension for quantum processing, mitigate vanishing gradients, and enhance trainability while improving the quantum circuit's flexibility in modeling complex decision policies for MO-TS. Numerical studies on IEEE 14-bus, 57-bus, and 118-bus systems demonstrate that the proposed algorithm achieves superior training stability and faster convergence with approximately 1% of the network parameters required by classical algorithms, highlighting its effectiveness, efficiency, and scalability. Furthermore, the practicality is validated through its stable convergence under three common quantum noise channels.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2025

Reinforcement Learning-Based Oscillation Dampening: Scaling Up Single-Agent Reinforcement Learning Algorithms to a 100-Autonomous-Vehicle Highway Field Operational Test

In this article, we explore the technical details of the reinforcement learning (RL) algorithms that were deployed in the largest field test of automated vehicles designed to smooth traffic flow in history as of 2023, uncovering the challenges and breakthroughs that come with developing RL controllers for automated vehicles. We delve into the fundamental concepts behind RL algorithms and their application in the context of self-driving cars, discussing the developmental process from simulation to deployment in detail, from designing simulators to reward function shaping. We present the results in both simulation and deployment, discussing the flow-smoothing benefits of the RL controller. From understanding the basics of Markov decision processes to exploring advanced techniques such as deep RL, our article offers a comprehensive overview and deep dive of the theoretical foundations and practical implementations driving this rapidly evolving field. We also showcase real-world case studies and alternative research projects that highlight the impact of RL controllers in revolutionizing autonomous driving. From tackling complex urban environments to dealing with unpredictable traffic scenarios, these intelligent controllers are pushing the boundaries of what automated vehicles can achieve. Furthermore, we examine the safety considerations and hardware-focused technical details surrounding deployment of RL controllers into automated vehicles. As these algorithms learn and evolve through interactions with the environment, ensuring their behavior aligns with safety standards becomes crucial. Here, we explore the methodologies and frameworks being developed to address these challenges, emphasizing the importance of building reliable control systems for automated vehicles.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Quick Talk: Resilient Communities, Maryland

Groundswell, Inc., in partnership with the Maryland Energy Administration (MEA), Ayika Solutions, the National Renewable Energy Lab (NREL), and the Baltimore Office of Sustainability, is advancing a vision for resilient, community-centered energy systems in Maryland. This effort aligns with Groundswell's commitment to a clean energy economy that equitably benefits all communities, closing gaps in representation, wages, and access to wealth generation, savings, and local jobs within a diverse energy supply chain. Project Vision and Objectives The project emphasizes "energy resilience," defined as a community’s capacity to anticipate, adapt to, and swiftly recover from power disruptions through inclusive planning and technical solutions. The approach centers on Resilience Hubs—designated community facilities equipped to support residents before, during, and after power-related disruptions. These hubs offer emergency resources, coordinate communications, and deliver essential services while reducing carbon emissions, ultimately enhancing residents' quality of life. Additionally, Resilience Hubs foster social connections, support disaster preparedness, and provide spaces for community programming and everyday use. The goals of Resilient Communities Maryland include: Developing a replicable framework for community energy resilience, applicable beyond Maryland. Defining community-led indicators of social impact and resilience. Establishing equitable decision-making processes in project design and implementation. Informing MEA of practical, community-based resilience strategies to shape future state initiatives. Approach Groundswell and its partners prioritize collaborative engagement with local stakeholders, ensuring that solutions are responsive to each community's unique needs. By integrating feedback from residents, local organizations, and government entities, the team has developed a holistic model for resilience that balances immediate benefits, such as improved emergency response, with long-term gains in social equity and environmental sustainability. Key Outcomes This initiative has produced valuable resources and insights, including: A Community Energy Resilience Framework: Designed to guide other communities in establishing energy resilience strategies tailored to their specific needs. Case Studies: Documenting best practices and lessons learned from pilot projects to serve as models for future resilience efforts. Community-Driven Impact Metrics: Highlighting the positive social impacts that resilience initiatives can bring, such as improved social connectedness and local decision-making power. Recommendations for MEA: Offering evidence-based insights to inform policies and funding opportunities that support resilient energy infrastructure in vulnerable communities. Next Steps Groundswell's final report will detail findings, including the potential for scaling resilience hubs across Maryland and other regions, with a focus on equitable access and community-driven outcomes. This project underscores the importance of aligning energy resilience with social equity, reinforcing community agency, and building systems that withstand and recover from future challenges effectively and sustainably.

14 SOLAR ENERGY↗
DOE OSTI2024

"Reinforcement Learning Based Oscillation Dampening: Scaling up Single-Agent RL algorithms to a 100 AV highway field operational test"

In this article, we explore the technical details of the reinforcement learning (RL) algorithms that were deployed in the largest field test of automated vehicles designed to smooth traffic flow in history as of 2023, uncovering the challenges and breakthroughs that come with developing RL controllers for automated vehicles. We delve into the fundamental concepts behind RL algorithms and their application in the context of self-driving cars, discussing the developmental process from simulation to deployment in detail, from designing simulators to reward function shaping. We present the results in both simulation and deployment, discussing the flow-smoothing benefits of the RL controller. From understanding the basics of Markov decision processes to exploring advanced techniques such as deep RL, our article offers a comprehensive overview and deep dive of the theoretical foundations and practical implementations driving this rapidly evolving field. We also showcase real-world case studies and alternative research projects that highlight the impact of RL controllers in revolutionizing autonomous driving. From tackling complex urban environments to dealing with unpredictable traffic scenarios, these intelligent controllers are pushing the boundaries of what automated vehicles can achieve. Furthermore, we examine the safety considerations and hardware-focused technical details surrounding deployment of RL controllers into automated vehicles. As these algorithms learn and evolve through interactions with the environment, ensuring their behavior aligns with safety standards becomes crucial. We explore the methodologies and frameworks being developed to address these challenges, emphasizing the importance of building reliable control systems for automated vehicles.

Jang, Kathy↗