로보틱스 연구일지

최근 VANP: Learning Where to See for Navigation with Self-Supervised Vision-Action Pre-Training라는 논문을 읽었다. 간략하게는 Navigation시 어디를 집중해서 볼지 제안한 논문이다. https://arxiv.org/abs/2403.08109 VANP: Learning Where to See for Navigation with Self-Supervised Vision-Action Pre-TrainingHumans excel at efficiently navigating through crowds without collision by focusing on specific visual regions relevant to navig..