Demonstrating Arena 5.0: A Photorealistic ROS2 Simulation Framework for Developing and Benchmarking Social Navigation

📅 2026-10-08
📈 Citations: 0
✨ Influential: 0
📄 PDF
🤖 AI Summary
This study addresses the challenges of low simulation fidelity, scarce benchmarks, and constrained scenario generation in social navigation research by proposing the Arena 5.0 framework. This work pioneers the seamless integration of NVIDIA Isaac Gym into the ROS2 platform, unifying environment generation, evaluation tools, and planner interfaces. By incorporating domain randomization techniques, the framework achieves high-fidelity simulation and customizable scenario generation while establishing a comprehensive multi-metric evaluation system. The project has been open-sourced, and user studies confirm that its usability and training efficiency are significantly improved over previous generations. Furthermore, Arena 5.0 demonstrates strong capability in supporting complex application scenarios, such as emergency rescue simulations.
📝 Abstract
Building upon the foundations laid by our previous work, this paper introduces Arena 5.0, the fifth iteration of our framework for robotics social navigation development and benchmarking. Arena 5.0 provides three main contributions: 1) The complete integration of NVIDIA Isaac Gym, enabling photorealistic simulations and more efficient training. It seamlessly incorporates Isaac Gym into the Arena platform, allowing the use of existing modules such as randomized environment generation, evaluation tools, ROS2 support, and the integration of planners, robot models, and APIs within Isaac Gym. 2) A comprehensive benchmark of state-of-the-art social navigation strategies, evaluated on a diverse set of generated and customized worlds and scenarios of varying difficulty levels. These benchmarks provide a detailed assessment of navigation planners using a wide range of social navigation metrics. 3) Extensive scenario generation and task planning modules for improved and customizable generation of social navigation scenarios, such as emergency and rescue situations. The platform's performance was evaluated by generating the aforementioned benchmark and through a comprehensive user study, demonstrating significant improvements in usability and efficiency compared to previous versions. Arena 5.0 is open source and available at https://github.com/Arena-Rosnav.
Problem

Research questions and friction points this paper is trying to address.

Social Navigation
Simulation Framework
Benchmarking
ROS2
Photorealistic Simulation
Innovation

Methods, ideas, or system contributions that make the work stand out.

Social Navigation
Photorealistic Simulation
Isaac Gym
ROS2
Benchmarking
🔎 Similar Papers
No similar papers found.
V
Volodymyr Shcherbyna
Technical University Berlin (TUB), Germany
L
Linh Kästner
National University of Singapore (NUS), Singapore; Technical University Berlin (TUB), Germany
D
Duc Anh Do
National University of Singapore (NUS), Singapore
H
Hoang Tung
Technical University Munich (TUM), Germany
H
Huu Giang Nguyen
Technical University Berlin (TUB), Germany
M
Maximilian Ho-Kyoung Schreff
Technical University Munich (TUM), Germany
T
Tim Seeger
Technical University Berlin (TUB), Germany
Eva Wiese
Eva Wiese
Technical University Berlin (TUB), Germany
A
Ahmed Martban
Technical University Berlin (TUB), Germany
H
Huajian Zeng
Technical University Munich (TUM), Germany
A
An Tran
National University of Singapore (NUS), Singapore
N
Nguyen Quoc Hung
Technical University Berlin (TUB), Germany
J
Jonas Kreutz
Technical University Berlin (TUB), Germany
V
Vu Thanh Lam
Technical University Berlin (TUB), Germany
T
Ton Manh Kien
Technical University Berlin (TUB), Germany
Harold Soh
Harold Soh
Associate Professor at National University of Singapore
Human Robot InteractionMachine LearningTactile PerceptionArtificial IntelligenceRobotics