RoboMemory: A Brain-inspired Multi-memory Agentic Framework for Interactive Environmental Learning in Physical Embodied Systems
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911539403948032 |
|---|---|
| author | Lei, Mingcong Cai, Honghao Yang, Yuyuan Wu, Yimou Ren, Jinke Cui, Zezhou Tan, Liangchen Hong, Junkun Hu, Gehan Zhu, Shuangyu Jiang, Shaohan Wang, Ge Tan, Junyuan Wan, Zhenglin Li, Zheng Li, Zhen Cui, Shuguang Zhao, Yiming Han, Yatong |
| author_facet | Lei, Mingcong Cai, Honghao Yang, Yuyuan Wu, Yimou Ren, Jinke Cui, Zezhou Tan, Liangchen Hong, Junkun Hu, Gehan Zhu, Shuangyu Jiang, Shaohan Wang, Ge Tan, Junyuan Wan, Zhenglin Li, Zheng Li, Zhen Cui, Shuguang Zhao, Yiming Han, Yatong |
| contents | Embodied intelligence aims to enable robots to learn, reason, and generalize robustly across complex real-world environments. However, existing approaches often struggle with partial observability, fragmented spatial reasoning, and inefficient integration of heterogeneous memories, limiting their capacity for long-horizon adaptation. To address this, we introduce RoboMemory, a brain-inspired framework that unifies Spatial, Temporal, Episodic, and Semantic memory within a parallelized architecture for efficient long-horizon planning and interactive learning. Its core innovations are a dynamic spatial knowledge graph for scalable, consistent memory updates and a closed-loop planner with a critic module for adaptive decision-making. Extensive experiments on EmbodiedBench show that RoboMemory, instantiated with Qwen2.5-VL-72B-Ins, improves the average success rate by 26.5% over its strong baseline and even surpasses the closed-source SOTA, Claude-3.5-Sonnet. Real-world trials further confirm its capability for cumulative learning, with performance consistently improving over repeated tasks. Our results position RoboMemory as a scalable foundation for memory-augmented embodied agents, bridging insights from cognitive neuroscience with practical robotic autonomy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_01415 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RoboMemory: A Brain-inspired Multi-memory Agentic Framework for Interactive Environmental Learning in Physical Embodied Systems Lei, Mingcong Cai, Honghao Yang, Yuyuan Wu, Yimou Ren, Jinke Cui, Zezhou Tan, Liangchen Hong, Junkun Hu, Gehan Zhu, Shuangyu Jiang, Shaohan Wang, Ge Tan, Junyuan Wan, Zhenglin Li, Zheng Li, Zhen Cui, Shuguang Zhao, Yiming Han, Yatong Robotics Artificial Intelligence Embodied intelligence aims to enable robots to learn, reason, and generalize robustly across complex real-world environments. However, existing approaches often struggle with partial observability, fragmented spatial reasoning, and inefficient integration of heterogeneous memories, limiting their capacity for long-horizon adaptation. To address this, we introduce RoboMemory, a brain-inspired framework that unifies Spatial, Temporal, Episodic, and Semantic memory within a parallelized architecture for efficient long-horizon planning and interactive learning. Its core innovations are a dynamic spatial knowledge graph for scalable, consistent memory updates and a closed-loop planner with a critic module for adaptive decision-making. Extensive experiments on EmbodiedBench show that RoboMemory, instantiated with Qwen2.5-VL-72B-Ins, improves the average success rate by 26.5% over its strong baseline and even surpasses the closed-source SOTA, Claude-3.5-Sonnet. Real-world trials further confirm its capability for cumulative learning, with performance consistently improving over repeated tasks. Our results position RoboMemory as a scalable foundation for memory-augmented embodied agents, bridging insights from cognitive neuroscience with practical robotic autonomy. |
| title | RoboMemory: A Brain-inspired Multi-memory Agentic Framework for Interactive Environmental Learning in Physical Embodied Systems |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2508.01415 |