Conflict resolution problem (CRP) plays a crucial role in the guarantee of safety. This paper formulates CRP as a multi-agent path planning problem which aims to find optimal paths for aircrafts. An algorithm named CCDG is proposed to tackle it based on cooperative coevolutionary (CC) with a dynamic grouping strategy for aircraft. CCDG makes aircraft divided into several equal sub-groups according to the dynamic grouping strategy. Each sub-group can adopt an evolutionary algorithm (EA) to optimize the aircrafts paths fully distributed and in parallel. Optimal solution is obtained through cooperation and coordination with all sub-populations. Empirical studies using extremely scenario adopted by previous research show that CCDG outperformed the existing approach (the fast GA), and the popular path planner that each aircraft uses an EA. Moreover, CCDG succeed to improve the airspace safety and reduce cost for CR.


    Access

    Access via TIB

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Conflict Resolution Based on Cooperative Coevolutionary with Dynamic Grouping Strategy for Multi-Aircraft


    Contributors:
    Liu, En-Yu (author) / Guan, Xiang-Min (author) / Zhang, Xue-jun (author) / Zeng, Jie (author)


    Publication date :

    2013


    Size :

    5 Seiten




    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English





    An Airborne Emergency Conflict Resolution against a Non-Cooperative Aircraft

    Kuo, V. / American Institute of Aeronautics and Astronautics | British Library Conference Proceedings | 2010


    Aircraft conflict resolution cataloguer

    Yao, Chu / Rusu, Adrian / Danick, Andrew et al. | IEEE | 2017


    Graph Reinforcement Learning for Multi-Aircraft Conflict Resolution

    Li, Yumeng / Zhang, Yunhe / Guo, Tong et al. | IEEE | 2024


    Intelligent resolution of cooperative conflict

    Li, L.-s. / Yu, H.-x. / Luo, J.-x. et al. | British Library Online Contents | 2000