基于差异的关系学习用于零样本目标视觉导航及直接模拟到现实的迁移
Difference-Based Relational Learning for Zero-Shot Object-Goal Visual Navigation With Direct Sim-to-Real Transfer
浏览论文内容
中文总结 AI 辅助
针对零样本目标视觉导航中模拟到现实差距的挑战，提出T-DRN，结合基于暹罗差异的特征提取器和双帧时间缓冲区，经实验验证其在AI2-THOR及物理轮式机器人上能提高零样本泛化能力，支持直接模拟到现实迁移。
中文摘要 AI 辅助
端到端深度强化学习用于零样本目标视觉导航仍受模拟到现实差距的挑战，尤其是物体外观变化和受限相机视野。本文提出一种时间差分关系网络（T-DRN）用于稳健的零样本模拟到现实迁移。它结合基于暹罗差异的特征提取器计算目标与观察物体的关系差异以产生独立于域的表示，还有双帧时间缓冲区在窄视野下保持短期物体连续性。在AI2-THOR中的大量实验表明T-DRN在成功率方面优于强基线提高了零样本泛化能力。此外，在物理轮式机器人上系统验证了T-DRN，证明其在实际传感和驱动约束下性能稳健，支持直接模拟到现实迁移的可行性。
英文摘要
End-to-end deep reinforcement learning (DRL) for zero-shot object-goal visual navigation remains challenged by the sim-to-real gap, particularly variations in object appearance and restricted camera field-of-view (FoV). This letter proposes a Temporal Difference-Relational Network (T-DRN) for robust zero-shot sim-to-real transfer. T-DRN combines a Siamese difference-based feature extractor, which computes relational difference between the target and observed objects to produce domain-independent representations, with a dual-frame temporal buffer that preserves short-term object continuity under narrow FoV. Extensive experiments in AI2-THOR demonstrate that T-DRN improves zero-shot generalization in terms of success rates over strong baselines. Furthermore, T-DRN is systematically validated on a physical wheeled robot, demonstrating robust performance under real sensing and actuation constraints and supporting the feasibility of direct sim-to-real transfer.
