NavBench: A Unified Robotics Benchmark for Reinforcement Learning-Based Autonomous Navigation

2025-05-20

AI Summary

Existing RL navigation benchmarks suffer from closed ecosystems and poor cross-domain generalization, hindering fair evaluation of multi-modal robots and sim-to-real transfer. To address this, we propose NavBench—the first cross-medium, multi-platform unified navigation benchmark, covering underwater, ground, aerial, and space domains. NavBench introduces a novel cross-medium task formulation, a modular evaluation framework, and a standardized RL interface built on IsaacLab, enabling comparable assessment across heterogeneous robotic platforms—including wheeled, jet-propelled, and underwater vehicles. We further enhance policy generalization via domain randomization and plug-and-play robot templates. Experiments demonstrate successful sim-to-real transfer to a satellite manipulator simulator, an unmanned surface vehicle, and a wheeled robot. The codebase is open-sourced to facilitate community-driven extension and benchmarking.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Computer Vision: Multi-modal Vision

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Systems and Infrastructure for Web, Mobile and WoT: Web applications in cross-disciplinary domains and verticals such as mixed reality, smart cities, and digital health
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Autonomous robots must navigate and operate in diverse environments, from terrestrial and aquatic settings to aerial and space domains. While Reinforcement Learning (RL) has shown promise in training policies for specific autonomous robots, existing benchmarks are often constrained to unique platforms, limiting generalization and fair comparisons across different mobility systems. In this paper, we present NavBench, a multi-domain benchmark for training and evaluating RL-based navigation policies across diverse robotic platforms and operational environments. Built on IsaacLab, our framework standardizes task definitions, enabling different robots to tackle various navigation challenges without the need for ad-hoc task redesigns or custom evaluation metrics. Our benchmark addresses three key challenges: (1) Unified cross-medium benchmarking, enabling direct evaluation of diverse actuation methods (thrusters, wheels, water-based propulsion) in realistic environments; (2) Scalable and modular design, facilitating seamless robot-task interchangeability and reproducible training pipelines; and (3) Robust sim-to-real validation, demonstrated through successful policy transfer to multiple real-world robots, including a satellite robotic simulator, an unmanned surface vessel, and a wheeled ground vehicle. By ensuring consistency between simulation and real-world deployment, NavBench simplifies the development of adaptable RL-based navigation strategies. Its modular design allows researchers to easily integrate custom robots and tasks by following the framework's predefined templates, making it accessible for a wide range of applications. Our code is publicly available at NavBench.

Problem

• Unified cross-medium benchmarking for diverse robotic platforms
• Scalable modular design enabling robot-task interchangeability
• Robust sim-to-real validation for real-world deployment

Innovation

• Unified cross-medium benchmarking for diverse actuation methods
• Scalable modular design enabling robot-task interchangeability
• Robust sim-to-real validation for real-world policy transfer