This paper presents a method for trajectory optimization to achieve road vehicle obstacle avoidance using convex optimization. It builds upon recent work on collision avoidance by automatic emergency lateral manoeuvre where the hierarchical control architecture consists of reference trajectory generator, feed forward controller, and feedback controller. The results of the paper concentrate on the most critical part of this control architecture, the reference trajectory generator, which must generate an optimal vehicle trajectory that is sufficiently aggressive yet feasible for the feed forward/feedback controller. More details of the resultant superior feed forward/feedback controller performance for a severe lane change derived from an ISO standard are reported elsewhere. There are three main conclusions. First, it is shown that, despite the presence of significant nonlinearities, it is possible to articulate the obstacle avoidance problem in a convex form if the optimization is allowed to converge to a solution over multiple passes. Specifying the problem in this way offers access to powerful and efficient numerical solvers that can take advantage of this convexity. Second, it is necessary to consider rotation of the vehicle relative to the fixed Earth axis if the reference trajectory is required to avoid obstacles specified in this axis system. It has been shown that, for even a simple periodic trajectory, the lateral displacement of the vehicle may be twice that which would be indicated if axis rotation is neglected. Finally, it is demonstrated that an optimal trajectory that accounts for the vehicle's changing velocity throughout the manoeuvre is superior to a previous analytical method that assumes constant speed. The procedure can generate a trajectory at higher vehicle speeds than would otherwise be possible.


    Zugriff

    Zugriff über TIB

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Trajectory generation for road vehicle obstacle avoidance using convex optimization


    Beteiligte:
    Bevan, G.P. (Autor:in) / Gollee, H. (Autor:in) / O'Reilly, J. (Autor:in)


    Erscheinungsdatum :

    2010


    Format / Umfang :

    19 Seiten, 11 Bilder, 28 Quellen




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Print


    Sprache :

    Englisch




    Trajectory generation for road vehicle obstacle avoidance using convex optimization

    Bevan, G P / Gollee, H / O'Reilly, J | SAGE Publications | 2010



    Trajectory generation for road vehicle obstacle avoidance using convex optimization

    Bevan,G.P. / Gollee,H. / OReilly,J. et al. | Kraftfahrwesen | 2010



    A Real-Time Obstacle-Avoidance Trajectory Planner for On-Road Autonomous Vehicle

    Li, Yishan / Lu, Xiong / Zeng, Dequan et al. | British Library Conference Proceedings | 2020