AstroMemBench：面向航天智能体记忆系统的基准测试发布
AstroMemBench: A Benchmark for Memory Systems in Embodied Spacecraft Agents
做航天自主智能体的团队可以关注这个基准。它把通信延迟、供电约束这些真实太空场景变成可量化的记忆测试，实测发现 Mem0 在会话重启后比 RAG 高 12.25 分，但平时两者差距不大。开源地址直接可用，做长程记忆系统设计的开发者可以拿来跑跑自己方案。
AstroMemBench 是一个针对航天器具身智能体记忆系统的基准，采用 Scenario-Challenge-Capability 设计方法，覆盖 16 个记忆子挑战，通过 140 个模板生成 2240 个实例，横跨行星表面作业、在轨服务、深空探索三类场景共九个任务。
基准用确定性环境对任务完成度、状态变化、安全性和资源使用进行评分，并设计了九种消融条件来检验记忆系统设计差异的影响。主实验比较六种配置，Mem0、Chain-of-Memory、LightMem 和 RAG 在单次会话内的平均得分均优于原生 Pi Agent，但三种记忆系统相比 RAG 没有明确优势。
在会话重启后的测试中，Mem0 比 RAG 高出 12.25 分，收益主要体现在指令保持和结果复用上，而在持续状态跟踪和能源预留方面优势有限。AstroMemBench 1.0 及评估代码已在 GitHub 开源，可用于具身航天智能体记忆系统的选型与开发。
AstroMemBench: A Benchmark for Memory Systems in Embodied Spacecraft Agents
Long-term spacecraft autonomy requires embodied agents to use operational history under communication delays, time-critical operations, environmental uncertainty, complex physical interactions, power and computational constraints, and safety and reliability requirements. We introduce AstroMemBench and a Scenario-Challenge-Capability design method that maps these requirements to executable memory tests. The benchmark covers 16 memory subchallenges through 140 templates and 2240 instances across nine tasks in planetary surface operations, on-orbit servicing, and deep-space exploration. A deterministic environment scores actions for task completion, state changes, safety, and resource use. Ablations under nine conditions show that the benchmark is sensitive to how accurately agents recover and use historical information and to differences in memory-system design. Main experiments compare six configurations. Mem0, Chain-of-Memory, LightMem, and retrieval-augmented generation (RAG) improve same-session mean scores over native Pi Agent, with no clear advantage of the three memory systems over RAG. After a session restart, Mem0 outperforms RAG by 12.25 points. Benefits favor instruction retention and result reuse over ongoing state tracking and energy reservation. AstroMemBench supports the design, development, and selection of memory systems for embodied spacecraft agents. AstroMemBench 1.0 and its evaluation code are publicly available: https://github.com/flyingView/AstroMemBench-v1.
本条情报归属 AITOP「论文」核心追踪栏目。在跨区域产业观察坐标中，该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线，反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括：AstroMemBench、Mem0、RAG、智能体，读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明：本文原始线索采编自「arXiv cs.AI」（@You Wu, Guhong Zhang, Hongwei Li, Xinyue Hong, Yuan Jia），经 AITOP 多维评分管线结构化提炼。