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.

翻译：具身智能旨在使机器人能够在复杂真实环境中稳健地学习、推理和泛化。然而，现有方法常受困于部分可观测性、碎片化空间推理以及异构记忆的低效整合，限制了其长期适应能力。为解决此问题，我们提出RoboMemory——一种受脑启发的框架，该框架将空间记忆、时间记忆、情景记忆和语义记忆统一于并行化架构中，以实现高效的长周期规划与交互式学习。其核心创新包括：用于可扩展一致性记忆更新的动态空间知识图谱，以及配备评论模块的闭环规划器以支持自适应决策。在EmbodiedBench上的大量实验表明，以Qwen2.5-VL-72B-Ins实例化的RoboMemory，相较于强基线模型，平均成功率提升26.5%，甚至超越闭源SOTA模型Claude-3.5-Sonnet。真实世界试验进一步证实了其累积学习能力——在重复任务中性能持续提升。我们的研究结果将RoboMemory定位为记忆增强型具身智能体的可扩展基础，桥接了认知神经科学洞察与实用机器人自主性之间的鸿沟。