Enhancing the NAO: Extending Capabilities of Legacy Robots for Long-Term Research

TL;DR

This work tackles the degradation of research value in unsupported legacy robots by proposing a platform-agnostic upgrade framework and demonstrating it on NAO. The Enhanced NAO combines a self-contained hardware pack (RGB-D and thermal cameras, Raspberry Pi 5, ReSpeaker, Hailo-8L) with a cloud-assisted yet streaming perception pipeline and GPT-4o-driven dialogue, enabling richer, open-ended interactions without adding latency. A pilot study with eight participants shows improved conversational quality and strong user preference for the Enhanced NAO over the NAO AI Edition. Collectively, the study provides a practical blueprint for extending the lifespan of legacy robots and highlights a path toward applying these upgrades to other platforms like Baxter and TurtleBot 2.

Abstract

Legacy (unsupported) robotic platforms often lose research utility when manufacturer support ends, preventing integration of modern sensing, speech, and interaction capabilities. We present the Enhanced NAO, a revitalized version of Aldebaran's NAO robot featuring upgraded beamforming microphones, RGB-D and thermal cameras, and additional compute resources in a fully self-contained package. This system combines cloud-based and local models for perception and dialogue, while preserving the NAO's expressive body and behaviors. In a pilot user study validating conversational performance, the Enhanced NAO delivered significantly higher conversational quality and elicited stronger user preference compared to the NAO AI Edition, without increasing response latency. The added visual and thermal sensing modalities established a foundation for future perception-driven interaction. Beyond this implementation, our framework provides a platform-agnostic strategy for extending the lifespan and research utility of legacy robots, ensuring they remain valuable tools for human-robot interaction.

Figures (7)

• Figure 1: Enhanced NAO robot with integrated sensing, computing, and power systems. A. Back isometric view showing the ReSpeaker 4 Mic Array, Raspberry Pi 5, Hailo-8L, and battery pack. B. Front view highlighting the Orbbec Gemini 2 L depth camera and Seek Thermal camera (S314SPX).
• Figure 2: Sample camera frames produced by the enhanced camera system. A) Processed RGB and B) depth frames from the Orbbec Gemini 2 L and (C) processed thermal frame from the Seek Thermal (S314SPX) camera.
• Figure 3: Audio processing pipeline, from audio input via the ReSpeaker to a response audio played back through the NAO speakers.
• Figure 4: A user engaging in a multi-turn conversation with the Enhanced NAO during the pilot validation study.
• Figure 5: Example of a participant conversation with the NAO AI and the Enhanced NAO.