Work

MultiDrive: A Co-Simulation Framework Bridging 2D and 3D Driving Simulation for AV Software Validation

Year: 2025

Type: conference-paper

Abstract: Scenario-based testing using simulations is a cornerstone of software validation for Autonomous Vehicles (AVs). So far, developers needed to choose between low-fidelity 2D simulators to explore the scenario space efficiently, and high-fidelity 3D simulators to study relevant scenarios in more detail, thus reducing testing costs while mitigating the sim-to-real gap. This paper presents a novel framework that leverages multi-agent co-simulation and procedural scenario generation to support scenario-based testing across low- and high-fidelity simulators for the development of motion planning algorithms. Our framework limits the effort required to transition scenarios between simulators and automates experiment execution, trajectory analysis, and visualization. Experiments with a reference motion planner demonstrate that our framework reveals discrepancies between the planner's intended and actual behavior, thus exposing weaknesses in planning assumptions under more realistic conditions. Our framework is available at: https://github.com/TUM-AVS/MultiDrive.

Source: IEEE International Conference on Intelligent Transportation Systems (ITSC)

DOI: 10.1109/itsc60802.2025.11423490

Authors: Marc Kaufeld, Korbinian Moller, Alessio Gambi, Paolo Arcaini, Johannes Betz

Institutions: Technical University of Munich, AIT Austrian Institute of Technology GmbH, National Institute of Informatics

Language: English

Topic: Autonomous Vehicle Technology and Safety

Subfield: Automotive Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: green

Funder: JST-Mirai Program