一句话：提出 ReCA 协同设计框架，从算法、系统、硬件三层加速协作具身智能体，实现 10.2× 加速与更高任务成功率。
协作具身系统集成感知、规划、动作与大语言模型推理，有望完成复杂长时多目标任务，但部署面临规划与通信延迟、可扩展性受限、低层执行敏感等效率问题。本文提出 ReCA，一个面向协作具身智能体系统加速的表征与协同设计框架，从算法、系统与硬件三层优化：算法层采用高效局部模型处理降低模型开销；系统层设计长短期记忆融合的双记忆结构、集中-去中心化层次协作规划及规划引导多步执行；硬件层采用高层规划 GPU 与低层规划加速器异构架构。在长时多目标任务上评估，ReCA 跨场景与规模泛化，相比最先进系统成功率提升 4.3%，加速 10.2 倍。
Cooperative embodied systems, where multiple agents collaborate through integrated perception, planning, action, and advanced reasoning powered by large language models (LLMs), show great potential for tackling complex, long-horizon, multi-objective tasks in real-world environments. Despite these algorithmic advancements, deploying embodied agents on current systems remains challenging due to prolonged planning and communication latency, limited scalability, and heightened sensitivity in low-level execution, all of which lead to significant system inefficiencies. This work proposes ReCA, a characterization and co-design framework dedicated to cooperative embodied agent system acceleration, aiming to enhance both task efficiency and system scalability. On the algorithm level, ReCA enables efficient local model processing to alleviate the substantial model costs. On the system level, ReCA presents a dual-memory structure with integrated long-term and short-term memory, hierarchical cooperative planning scheme with centralized and decentralized cooperation, and planning-guided multi-step execution for highly efficient and scalable cooperative embodied agent computation. On the hardware level, ReCA employs a heterogeneous hardware system with high-level planning GPU subsystem and low-level planning accelerator subsystem to ensure efficient and robust task execution. Evaluated across long-horizon multi-objective tasks, ReCA generalizes across application scenarios and system scales, achieving a 4.3% increase in successful missions with 10.2× speedup compared to the state-of-the-art cooperative embodied autonomous agent systems.
@inproceedings{WanDIQJZKRR25,
title = {ReCA: Integrated Acceleration for Real-Time and Efficient Cooperative Embodied Autonomous Agents},
author = {Zishen Wan and Yuhang Du and Mohamed Ibrahim and Jiayi Qian and Jason Jabbour and Yang (Katie) Zhao and Tushar Krishna and Arijit Raychowdhury and Vijay Janapa Reddi},
booktitle = {ASPLOS},
year = {2025},
doi = {10.1145/3676641.3716016}
}