郭兴, 李擎, 姚其家, 鲁小雅. 异构无人系统协同控制研究进展[J]. 工程科学学报. DOI: 10.13374/j.issn2095-9389.2024.01.12.001
引用本文: 郭兴, 李擎, 姚其家, 鲁小雅. 异构无人系统协同控制研究进展[J]. 工程科学学报. DOI: 10.13374/j.issn2095-9389.2024.01.12.001
Research progress for cooperative control of heterogeneous unmanned systems[J]. Chinese Journal of Engineering. DOI: 10.13374/j.issn2095-9389.2024.01.12.001
Citation: Research progress for cooperative control of heterogeneous unmanned systems[J]. Chinese Journal of Engineering. DOI: 10.13374/j.issn2095-9389.2024.01.12.001

异构无人系统协同控制研究进展

Research progress for cooperative control of heterogeneous unmanned systems

  • 摘要: 面对复杂化、多样化和立体化的任务需求,异构无人系统将空、地、海跨域优势互补,最大化提升系统各方面能力,为未来协同作战和智慧民生提供有力的技术保障。首先,本文梳理了国内外在跨域协同领域出台的推动性文件,介绍了异构无人系统在军用、科研和民用等三方面的实际工程应用;其次,本文仅针对异构无人系统的关键技术之一的协同控制,从控制理论方法和控制工程应用出发,分别阐述了异构无人系统协同控制在一致性、轨迹跟踪、编队-合围和不同通信条件等四方面的最新进展。其中,由于一致性和轨迹跟踪存在上层控制理论性和下层控制可行性的区别,进一步从滑模控制、自适应控制、反步控制和自适应动态规划等四方面对轨迹跟踪研究展开阐述;然后,为了更好地促进异构无人系统协同控制在实际中的应用,通过总结前人相关工作,讨论了其在多约束控制与实时性、多任务切换控制和跨域通信下稳定性等三方面亟需解决的技术瓶颈。最后,综合异构无人系统协同控制研究现状和实际需求来看,指出了其在深度强化学习和分布式博弈、人机交互和反群体智能等三方面的发展趋势。

     

    Abstract: Considering complex, diverse and stereoscopic task requirements, heterogeneous unmanned systems (HUS) integrate across-domain advantages of air, ground and sea, respectively, which maximizes the overall capabilities of the system and provides strong technical support for future collaborative operations and smart livelihoods. Firstly, this paper reviews the promotional documents issued in the field of cross-domain collaboration both domestically and internationally, and introduces the practical engineering applications of HUS in military, scientific research and civilian fields. Secondly, this paper only focuses on collaborative control, one of the key technologies of HUS. And based on control theory and control engineering, the latest research progress in four aspects: consensus, trajectory tracking, formation-containment and different communication conditions, was elaborated. Among them, due to the difference between the theoretical nature of upper level control and the feasibility of lower level control in consensus and trajectory tracking, further research on trajectory tracking was shown from four aspects: sliding mode control, adaptive control, backstepping control and adaptive dynamic programming. Thirdly, in order to better promote the research and the practical application of HUS collaborative control, the technical?bottlenecks that urgently need to be addressed in three aspects: multiple constraints and real-time problem, multi-task switching control problem and stability under cross-domain communication, were discussed by summarizing the previous relevant work. Finally, based on the current research status and practical needs of HUS collaborative control, its development directions will tend towards three aspects: deep reinforcement learning and distributed games, human-machine interaction and anti-swarm intelligence.

     

/

返回文章
返回