BCNav: Bearing-Conditioned Depth Policies for Sound Source Navigation

2026-09-29

AI Summary

This study addresses the challenges of acoustic source navigation, specifically the domain gap caused by low-fidelity acoustic simulation and the kinematic mismatch arising from discrete action spaces. To overcome these limitations, this work proposes a novel framework that decouples acoustic processing from navigation policies. By leveraging direction-of-arrival (DOA) estimation to extract azimuth angles alongside depth images, an imitation learning policy is trained to output continuous velocity commands. Furthermore, robustness is enhanced through calibrated noise injection, enabling deployment without real audio data or fine-tuning. The proposed approach is validated on both simulated and physical robotic platforms, demonstrating reliable performance in unseen environments. Ultimately, this method significantly narrows the sim-to-real performance gap, achieving efficient and autonomous acoustic source navigation.

Abstract

The ability to navigate toward sound sources extends a robot's reach beyond its visual field, enabling response to auditory events in unknown environments. To equip robots with this capability, existing methods couple acoustic and visual information through joint audio-visual learning in acoustic simulators. However, acoustic simulation is both low-fidelity and expensive, producing a domain gap that prevents reliable real-world deployment, while the discrete action spaces inherited from grid-based simulators introduce an additional kinematic gap on physical robots. To alleviate these issues, we propose BCNav, a decoupled framework that separates the acoustic module from the learned navigation policy using direction-of-arrival (DOA) estimation: an estimator provides a scalar bearing to the sound source, so the navigation policy only processes depth images and a bearing angle, two inputs whose domain gaps are well characterized. We collect shortest-path demonstrations with calibrated bearing noise injection and train the policy via imitation learning to output continuous velocity commands directly executable on ground robots. We demonstrate the method in simulation and on a physical robot, navigating unknown environments without any acoustic fine-tuning, prior mapping, or real-world audio data collection. Code is available at https://github.com/york1to/bcnav.

Problem

• Sound Source Navigation
• Domain Gap
• Kinematic Gap
• Acoustic Simulation
• Robot Navigation

Innovation

• Sound Source Navigation
• Decoupled Framework
• Direction-of-Arrival Estimation
• Imitation Learning
• Sim-to-Real Transfer