Micropayments, involving low-value transactions (e.g., fractions of a euro/dollar), are critical for unlocking granular digital services. In this paper we present a novel highlevel architecture integrating artificial intelligence (AI) agents with the Bitcoin Lightning Network (LN) to enable efficient micropayments for electric vehicle (EV) charging and peer-to-peer energy trading. The proposed architecture leverages the ultra fast and low-cost nature of the LN to enforce trustless payments upon verified energy delivery. AI agents embedded in EVs and charging stations autonomously negotiate dynamic pricing and energy allocation using reinforcement learning (RL) approaches, optimizing grid load balancing and enhancing profitability compared to on-chain methods. Based on a comprehensive use case involving EV owners, operators and energy providers, we demonstrate the system’s viability, supported by a prototype implementation on the LN Testnet. Results show a $\mathbf{9 8. 2 \%}$ success rate for micropayments during simulated charging sessions, with AI agents reducing latency by prioritizing high-liquidity payment channels.

Pricing decisions in Supply Chain (SC) management play a crucial role in profit maximization. However, SC dynamics are increasingly complex and optimizing pricing-related decisions remains challenging due to centralized models that cause inefficiencies and slow adaptation. Trust issues between manufacturers and retailers, often driven by information asymmetry and data manipulation risks, also hinder proper SC coordination. In this paper we present a novel decentralized framework for dynamic pricing in SCs by integrating multi-agent reinforcement learning (MARL) with blockchain technology. We model the SC as a Stackelberg game where multiple manufacturers, acting as leaders, independently set wholesale prices and a single retailer, as the follower, adjusts retail prices to maximize profit. Each agent employs RL algorithms to iteratively learn optimal pricing strategies in response to evolving market conditions and competitor actions. The integration of a permissioned blockchain ensures that all pricing decisions, transactions and smart contract executions are immutably recorded, thus enhancing data integrity, security and transparency. Experimental evaluations under diverse market scenarios demonstrate that the proposed framework improves convergence rates, profit margins and system robustness. Our results highlight the potential of combining decentralized decision-making with tamper-proof ledger technology to foster trust and efficiency in complex SC ecosystems.

The emergence of Industry 4.0 technologies has significantly transformed supply chain operations, particularly through the deployment of Autonomous Guided Vehicles (AGVs) in logistics and manufacturing settings. Integrating Industrial Internet of Things (IIoT) devices with Artificial Intelligence (AI) has enhanced AGV autonomy by enabling real-time data-driven decision-making. However, challenges related to cybersecurity, data synchronization, and scalability, still persist in cyber-physical (CPS) manufacturing systems. Blockchain technology offers a prominent pathway towards ensuring data integrity, decentralization, and security, but its adoption in AGV applications remains limited due to scalability, latency, and computational constraints. To address this gap, this paper proposes a hybrid, blockchain-centric architecture that leverages the synergistic potential of Reinforcement Learning (RL) methods in multi-agent, collaborative AGVs. The architecture is designed to be scalable, interoperable, and resilient to cyber threats, making it suitable for a wide range of industrial applications.

Maintaining balance in the electrical grid is crucial for assuring stability, minimizing energy waste and integrating renewable energy sources. This paper introduces a multi-agent reinforcement learning (MARL) framework whereby Renewable Energy Producers (REPs), BTC Miners, and Energy Manager Agents (EMAs) interact dynamically to stabilize the grid. The system utilizes energy market data and reinforcement learning algorithms to motivate miners to modify their power usage habits according to the prevailing grid circumstances. Miners, as energy consumers within the system, enable the absorption of excess renewable energy during low-demand periods and reduce use during peak times to relieve grid stress. We use the proposed MARL method in a simulated setting to assess grid stability, profitability and energy efficiency. The findings indicate that BTC mining can markedly reduce the volatility of the grid, improve the profitability of the miner and facilitate the incorporation of renewable energy.

We consider the problem of estimating continuous-time autoregressive (CAR) processes from discrete-time noisy observations. This can be done within a Bayesian framework using Markov chain Monte Carlo (MCMC) methods. Existing methods include the standard random walk Metropolis algorithm. On the other hand, least-squares (LS) algorithms exist where derivatives are approximated by differences and parameter estimation is done in a least-squares manner. In this paper, we incorporate the LS estimation into the MCMC framework to develop a new MCMC algorithm. This new algorithm is combined with the standard Metropolis algorithm and is found to improve performance compared to the standard MCMC algorithm. Simulation results are presented to support our findings.