Semantic Glitch: Agency and Artistry in an Autonomous Pixel Cloud

TL;DR

This paper introduces 'Semantic Glitch,' a lo-fi robotic art installation that uses semantic understanding from large language models for navigation and personality creation, emphasizing artistic expression over precision.

Contribution

It presents a novel autonomous pipeline that relies solely on semantic understanding for navigation and character creation, bypassing traditional sensors like LiDAR and SLAM.

Findings

Emergent behaviors such as landmark-based navigation.

Robustness of the framework for creating distinct personas.

Unpredictable, character-driven behaviors from limited proprioception.

Abstract

While mainstream robotics pursues metric precision and flawless performance, this paper explores the creative potential of a deliberately "lo-fi" approach. We present the "Semantic Glitch," a soft flying robotic art installation whose physical form, a 3D pixel style cloud, is a "physical glitch" derived from digital archaeology. We detail a novel autonomous pipeline that rejects conventional sensors like LiDAR and SLAM, relying solely on the qualitative, semantic understanding of a Multimodal Large Language Model to navigate. By authoring a bio-inspired personality for the robot through a natural language prompt, we create a "narrative mind" that complements the "weak," historically, loaded body. Our analysis begins with a 13-minute autonomous flight log, and a follow-up study statistically validates the framework's robustness for authoring quantifiably distinct personas. The combined analysis reveals emergent behaviors, from landmark-based navigation to a compelling "plan to execution" gap, and a character whose unpredictable, plausible behavior stems from a lack of precise proprioception. This demonstrates a lo-fi framework for creating imperfect companions whose success is measured in character over efficiency.