In Unmanned Aeronautical Ad hoc Networks (UAANETs), it is challenging for routing protocols such as ad hoc on-demand distance vector (AODV) routing and improved protocols to provide reliable communications due to limited wireless resources and high mobility. This paper proposes a modified AODV protocol applying multi-agent Q-learning(Modified-QLAODV) based on mobility prediction and load sensing. This protocol is a distributed multi-agent reinforcement routing strategy, which employs reinforcement learning algorithm. It introduces mobility prediction and node relative load to infer link state on the purpose of enhancing robustness of UAANETs. In the learning strategy, mobility prediction based on Kalman filter algorithm combined with Gaussian filter algorithm, and load sensing will have impacts on Q-values. The destination obtains routes with different Q-values and returns the one with the maximum average Q-value to the source, which realizes to select the optimal route from a global view. The simulation results demonstrate that Modified-QLAODV can outperform the Q-learning routing protocol(QLAODV) and AODV on network performance in UAANETs.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    A Modified Multi-Agent Reinforcement Learning Protocol Based on Prediction for UAANETs


    Contributors:
    Li, Chao (author) / Liu, Jing (author)


    Publication date :

    2020-11-01


    Size :

    1988436 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Underwater Acoustic MAC Protocol for Multi-Objective Optimization Based on Multi-Agent Reinforcement Learning

    Jinfang Jiang / Yiling Dong / Guangjie Han et al. | DOAJ | 2025

    Free access

    Collision prediction method and system based on multi-agent reinforcement learning

    YU ZHENG / LIN RONGSHENG / YU YONGQING et al. | European Patent Office | 2024

    Free access


    Multi-agent Cooperative Search based on Reinforcement Learning

    Sun, Yinjiang / Zhang, Rui / Liang, Wenbao et al. | IEEE | 2020


    Experience generalization for multi-agent reinforcement learning

    Pegoraro, Renê / Costa, AHR / Ribeiro, CHC | BASE | 2001

    Free access