Replicating the behaviour of electric vehicle drivers using an agent-based reinforcement learning model

2025-07-28

AI Summary

This study addresses three key challenges in modeling private electric vehicle (EV) driver charging behavior: (1) difficulty in scaling behavioral models across large geographic regions, (2) insufficient representation of bounded rationality and adaptive decision-making, and (3) lack of quantitative foundations for identifying “charging deserts.” To this end, we propose the first large-scale, multi-stage reinforcement learning agent model tailored to long-distance travel scenarios. The model integrates real-world driving trajectories with granular charging infrastructure data and employs staged policy training to accurately capture drivers’ heuristic charging decisions. Experimental results demonstrate high-fidelity replication of empirical charging patterns and a substantial improvement in low-state-of-charge risk prediction accuracy. Furthermore, our framework systematically identifies highway junctions and urban peripheries as critical charging deserts—the first such quantitative identification—thereby providing actionable evidence that strategic deployment of fast-charging hubs in these zones effectively alleviates charging stress during long-haul EV travel.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Search and Optimization: Learning to Search
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: The sharing economy

Abstract

Despite the rapid expansion of electric vehicle (EV) charging networks, questions remain about their efficiency in meeting the growing needs of EV drivers. Previous simulation-based approaches, which rely on static behavioural rules, have struggled to capture the adaptive behaviours of human drivers. Although reinforcement learning has been introduced in EV simulation studies, its application has primarily focused on optimising fleet operations rather than modelling private drivers who make independent charging decisions. Additionally, long-distance travel remains a primary concern for EV drivers. However, existing simulation studies rarely explore charging behaviour over large geographical scales. To address these gaps, we propose a multi-stage reinforcement learning framework that simulates EV charging demand across large geographical areas. We validate the model against real-world data, and identify the training stage that most closely reflects actual driver behaviour, which captures both the adaptive behaviours and bounded rationality of private drivers. Based on the simulation results, we also identify critical 'charging deserts' where EV drivers consistently have low state of charge. Our findings also highlight recent policy shifts toward expanding rapid charging hubs along motorway corridors and city boundaries to meet the demand from long-distance trips.

Problem

• Simulating adaptive EV driver behaviors using reinforcement learning
• Modeling private EV drivers' independent charging decisions
• Addressing long-distance charging needs across large geographical areas

Innovation

• Agent-based reinforcement learning for EV drivers
• Multi-stage framework for large geographical areas
• Identifying charging deserts using real-world data