🤖 AI Summary
This work addresses the lack of reliable passive visual communication in drone swarms operating in radio-constrained or interference-prone environments. The authors propose a motion-trajectory-based visual communication method, wherein sender drones execute dynamically feasible, modular planar trajectories as “visual sign language.” Receiver drones decode these signals through pose estimation and a custom-designed 3DTrajDecoder neural network, which performs trajectory classification, segmentation, and geometric parameter regression to enable non-active information exchange. The system integrates online procedural trajectory generation and is validated through a combined simulation-to-reality approach. Experiments demonstrate consistent performance in both simulated and real-world settings, delineate the operational domain boundaries, and confirm—via ablation studies—the architectural soundness and robustness of the proposed framework.
📝 Abstract
In stealth-constrained swarm robotics, visual communication provides a critical alternative to active radio transmissions, which might be jammed. This research investigates motion-based communication for non-active information exchange, utilizing modular, dynamically feasible planar trajectories as visual cues. On the receiver drone end, a pose estimator tracks the transmitting drone's pose, feeding it into our custom 3DTrajDecoder. The decoder is designed to classify and segment the spatiotemporal sequence while simultaneously regressing its size and normal vector. To robustly train the decoder on both communicative and non-communicative trajectories, we developed a configurable online procedural generation pipeline. We validate our system through real-world testing and simulation to define its operating domain, supported by an extensive ablation study detailing our architectural choices and system limitations.