深度が鍵：ロバスト自律エージェントのためのマルチモーダルRGB-D知覚
Depth Matters: Multimodal RGB-D Perception for Robust Autonomous Agents
RGB画像に深度情報を加えることで、自律エージェントのステアリング予測が大幅に向上することを示し、RGB-D特徴を融合した軽量リカレントコントローラを実機に展開して障害物回避を実現した。
著者: Mihaela-Larisa Clement, Mónika Farsang, Felix Resch, Mihai-Teodor Stanusoiu, Radu Grosu
分類: cs.RO, cs.CV, cs.LG
原文アブストラクト
Autonomous agents that rely purely on perception to make real-time control decisions require efficient and robust architectures. In this work, we demonstrate that augmenting RGB input with depth information significantly enhances our agents' ability to predict steering commands compared to using RGB alone. We benchmark lightweight recurrent controllers that leverage the fused RGB-D features for sequential decision-making. To train our models, we collect high-quality data using a small-scale autonomous car controlled by an expert driver via a physical steering wheel, capturing varying levels of steering difficulty. Our models were successfully deployed on real hardware and inherently avoided dynamic and static obstacles, under out-of-distribution conditions. Specifically, our findings reveal that the early fusion of depth data results in a highly robust controller, which remains effective even with frame drops and increased noise levels, without compromising the network's focus on the task.
