移动机器人路径规划中ROS2中间件性能的研究综述
A Review of the Performance of ROS2 Middleware in Mobile Robot Path Planning
DOI: 10.12677/csa.2025.158197, PDF,    科研立项经费支持
作者: 赵 鹏, 朱克佳:广州软件学院电子信息与控制工程学院,广东 广州
关键词: 中间件ROS2DDSZenoh路径规划Middleware ROS2 DDS Zenoh Path Planning
摘要: 随着移动机器人在工业自动化、特种作业及智能服务领域的广泛应用,其路径规划能力越来越依赖机器人操作系统ROS2的通信性能。ROS2通过去中心化架构与数据分发服务中间件显著提升了系统可靠性,但动态复杂环境中路径规划对通信延迟、带宽及稳定性的严苛要求,使中间件性能成为影响实时规划精度的关键瓶颈。近年来针对DDS、Zenoh等中间件的优化研究大量涌现,但仍缺乏对多协议多场景性能指标的跨维度系统性总结。本文综述了近年来核心研究成果,深入剖析了ROS2中间件从通用架构向场景定制化设计的转型趋势;论证了路径规划与通信服务质量的深度耦合机制;总结了嵌入式协同与边缘智能的系统级优化机制。文末提出了目前研究还存在的一些问题。
Abstract: With the wide application of mobile robots in industrial automation, special operations and intelligent services, their path planning capabilities are increasingly dependent on the communication performance of the Robot Operating System (ROS2). ROS2 has significantly enhanced system reliability through a decentralized architecture and Data Distribution Service (DDS) middleware. However, the strict requirements for communication delay, bandwidth and stability in dynamic and complex environments for path planning have made middleware performance a key bottleneck affecting real-time planning accuracy. In recent years, a large number of optimization studies on middleware such as DDS and Zenoh have emerged, but there is still a lack of cross-dimensional systematic summaries of performance indicators for multiple protocols and scenarios. This paper reviews the core research achievements in recent years, deeply analyzes the transformation trend of ROS2 middleware from a general architecture to scenario-customized design; demonstrates the deep coupling mechanism between path planning and communication quality of service; and summarizes the system-level optimization mechanisms of embedded collaboration and edge intelligence. At the end of the paper, some existing problems in current research are proposed.
文章引用:赵鹏, 朱克佳. 移动机器人路径规划中ROS2中间件性能的研究综述[J]. 计算机科学与应用, 2025, 15(8): 50-57. https://doi.org/10.12677/csa.2025.158197

参考文献

[1] Macenski, S., Foote, T., Gerkey, B., Lalancette, C. and Woodall, W. (2022) Robot Operating System 2: Design, Architecture, and Uses in the Wild. Science Robotics, 7, eabm6074. [Google Scholar] [CrossRef] [PubMed]
[2] Chovet, L.P., Garcia, G.M., Bera, A., Richard, A., Yoshida, K. and Olivares-Mendez, M.A. (2025) Performance Comparison of ROS2 Middlewares for Multi-Robot Mesh Networks in Planetary Exploration. Journal of Intelligent & Robotic Systems, 111, Article No. 18. [Google Scholar] [CrossRef
[3] Eclipse Zenoh (2022) Pub/Sub in Zenoh.
https://zenoh.io/docs/overview/zenoh-in-action/
[4] Corsaro, A., Cominardi, L., Hecart, O., Baldoni, G., Avital, J.E.P., Loudet, J., et al. (2023) Zenoh: Unifying Communication, Storage and Computation from the Cloud to the Microcontroller. 2023 26th Euromicro Conference on Digital System Design (DSD), Golem, 6-8 September 2023, 422-428. [Google Scholar] [CrossRef
[5] Zhang, J., Yu, X., Ha, S., Peña Queralta, J. and Westerlund, T. (2024) Comparison of Middlewares in Edge-to-Edge and Edge-to-Cloud Communication for Distributed ROS 2 Systems. Journal of Intelligent & Robotic Systems, 110, Article No. 162. [Google Scholar] [CrossRef
[6] 高华. ROS2的RTOS支持扩展研究与实现[D]: [硕士学位论文]. 北京: 北京邮电大学, 2020.
[7] 董利邦. 面向ROS2的RT-Smart通信中间件研究与实现[D]: [硕士学位论文]. 海口: 海南大学, 2023.
[8] 李冰鑫. 智慧机器人路径规划算法研究与应用[D]: [硕士学位论文]. 青岛: 青岛科技大学, 2024.
[9] 徐永成. 基于深度强化学习的移动机器人路径规划方法研究[D]: [硕士学位论文]. 大连: 大连理工大学, 2024.
[10] Maruyama, Y., Kato, S. and Azumi, T. (2016) Exploring the Performance of ROS2. Proceedings of the 13th International Conference on Embedded Software, New York, 1-7 October 2016, 1-10. [Google Scholar] [CrossRef
[11] eProsima (2022) Fast DDS vs Cyclone DDS.
https://www.eprosima.com/developer-resources/performance/fast-dds-vs-cyclone-dds-performance
[12] Bai, Z., Pang, H., He, Z., Zhao, B. and Wang, T. (2024) Path Planning of Autonomous Mobile Robot in Comprehensive Unknown Environment Using Deep Reinforcement Learning. IEEE Internet of Things Journal, 11, 22153-22166. [Google Scholar] [CrossRef
[13] Zheng, H., et al. (2025) Embodied Escaping: End-to-End Reinforcement Learning for Robot Navigation in Narrow Environment. Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). arXiv: 2503.03208.
[14] 武小年, 奚玉昂, 张润莲. DEM中基于遗传与蚁群的混合路径规划算法[J]. 计算机应用研究, 2020, 37(9): 2694-2697.
[15] 胡波, 侯琳. 双层随机机会约束规划的通信干扰任务分配优化仿真[J]. 计算机工程与应用 2014, 39(11): 59-63.
[16] 李纯艳, 晁永生, 陈帅, 等. 基于改进麻雀搜索算法的机器人能耗最优轨迹规划[J]. 组合机床与自动化加工技术, 2022(6): 180-182+187.
[17] 颜双权, 胥建成. 工业机器人复杂B样条曲线轨迹控制精度补偿[J]. 机械制造与自动化, 2023, 52(5): 32-35.
[18] 朱涵宗. 基于分布式模型预测控制算法的多移动机器人编队[D]: [硕士学位论文]. 武汉: 湖北工业大学, 2023.
[19] 章昌仲. 面向智能移动机器人的混合操作系统平台设计与实现[D]: [硕士学位论文]. 杭州: 浙江大学, 2020.
[20] 郭兴, 李擎, 姚其家, 等. 异构无人系统协同控制研究进展[J]. 工程科学学报, 2025, 47(1): 66-78.
[21] 王昕媛. 面向精准农业的异构无人系统协同方法研究[D]: [硕士学位论文]. 沈阳: 沈阳理工大学, 2022.
[22] 李宏. 轮式移动机器人的路径规划与跟踪控制研究[D]: [硕士学位论文]. 秦皇岛: 燕山大学, 2023.
[23] 王文重. 基于ROS2的分布式智能边缘云平台: WenKe [D]: [硕士学位论文]. 上海: 华东师范大学, 2024.
[24] López Escobar, J.J., Díaz-Redondo, R.P. and Gil-Castiñeira, F. (2024) Unleashing the Power of Decentralized Serverless IoT Dataflow Architecture for the Cloud-to-Edge Continuum: A Performance Comparison. Annals of Telecommunications, 79, 135-148. [Google Scholar] [CrossRef