CARJAN: Agent-Based Generation and Simulation of Traffic Scenarios with AJAN

2025-08-29

AI Summary

To address the challenges of modeling interactive agents and insufficient behavioral intelligence in urban traffic multi-agent simulation, this paper proposes an AJAN-CARLA integrated framework. It leverages the AJAN multi-agent engineering framework and the CARLA simulator to enable semi-automated generation and dynamic co-simulation of heterogeneous agents—including pedestrians, cyclists, and autonomous vehicles. Crucially, it introduces, for the first time in traffic simulation, a SPARQL-query-driven behavior tree decision mechanism, enabling knowledge-graph-guided, semantically rich behavioral control. A dedicated visual scenario editor is also developed to support interactive scenario construction and management. Experimental evaluation demonstrates significant improvements in scenario diversity, behavioral realism, and modeling efficiency. The framework establishes a high-fidelity, scalable virtual validation environment for autonomous driving system testing.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

User-friendly modeling and virtual simulation of urban traffic scenarios with different types of interacting agents such as pedestrians, cyclists and autonomous vehicles remains a challenge. We present CARJAN, a novel tool for semi-automated generation and simulation of such scenarios based on the multi-agent engineering framework AJAN and the driving simulator CARLA. CARJAN provides a visual user interface for the modeling, storage and maintenance of traffic scenario layouts, and leverages SPARQL Behavior Tree-based decision-making and interactions for agents in dynamic scenario simulations in CARLA. CARJAN provides a first integrated approach for interactive, intelligent agent-based generation and simulation of virtual traffic scenarios in CARLA.

Problem

• User-friendly modeling of urban traffic scenarios with diverse agents
• Semi-automated generation and simulation of interactive traffic scenarios
• Integrated approach for intelligent agent-based traffic simulation in CARLA

Innovation

• Agent-based traffic scenario generation
• SPARQL Behavior Tree decision-making
• Integrated CARLA simulation platform