Collision avoidance is a crucial task in vision-guided autonomous navigation. Traditional solutions tend to be computationally expensive and difficult to adapt to new environments. In this work, we propose a novel collision avoidance solution for autonomous drones. Formulated under a deep reinforcement learning framework, our model relies on a pair of margin reward functions to ensure the drones fly smoothly while greatly reducing the chance of collision. Additional reward functions are designed to attract the drones to fly towards their destinations, as well as to follow predefined routes. Experiments using indoor simulation environments demonstrate the effectiveness of our overall design and the individual components.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Vision-guided Collision Avoidance Through Deep Reinforcement Learning


    Contributors:
    Song, Sirui (author) / Zhang, Yuanhang (author) / Qin, Xi (author) / Saunders, Kirk (author) / Liu, Jundong (author)


    Publication date :

    2021-08-16


    Size :

    20107593 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Pedestrian Collision Avoidance Using Deep Reinforcement Learning

    Rafiei, Alireza / Fasakhodi, Amirhossein Oliaei / Hajati, Farshid | Springer Verlag | 2022


    Vision-based Distributed Multi-UAV Collision Avoidance via Deep Reinforcement Learning for Navigation

    Huang, Huaxing / Zhu, Guijie / Fan, Zhun et al. | ArXiv | 2022

    Free access



    Towards monocular vision based obstacle avoidance through deep reinforcement learning

    Xie, L / Wang, S / Markham, A et al. | BASE | 2020

    Free access