The invention discloses a motion planning method based on a safe trajectory tree network. The method comprises the following steps: generating a trajectory tree capable of representing discretization feasible spaces of all traffic participants through a TTNET network; depth feature codes, bounding box features of the autonomous vehicle and traffic participants around the autonomous vehicle and depth feature codes of local reference lines are obtained through multi-layer perceptron (MLP) coding; the depth feature codes are cascaded and then sent to a spatial Transform encoder, and through sufficient feature fusion, an overall interactive code is output; and inputting the overall interactive codes into the trajectory prediction assistance and safety trajectory tree network model, and outputting a final motion planning trajectory. Compared with an existing method, the driving score of the TTNET network on the CARLA Town05 closed-loop test benchmark is improved by 8.3%, and the route completion degree is improved by 39.2%; on the basis of CARLA Longest6, the traffic rule adherence score of the TTNET network is improved by 10.6%. Meanwhile, the TTNET network also has a reasoning speed of at least 28 Hz, which is 1.5 times faster than that of the best method.

    本发明公开了一种基于安全轨迹树网络的运动规划方法,包括:通过TTNET网络生成能够表示所有交通参与者的离散化可行空间的轨迹树;通过多层感知器MLP编码得到深度特征编码,自动驾驶汽车及其周围交通参与者的边界框特征和局部参考线的深度特征编码;这些深度特征编码之间经过级联后被送入空间Transformer编码器中,经过充分的特征融合,输出整体交互编码;将整体交互编码输入轨迹预测辅助与安全轨迹树网络模型,输出最终的运动规划轨迹。相较于现有方法,TTNET网络在CARLA Town05闭环测试基准上的驾驶分数提高了8.3%,路线完成度上提高了39.2%;在CARLA Longest6基准上,TTNET网络的遵守交通规则的分数提高了10.6%。同时,TTNET网络还具有至少28赫兹的推理速度,比最好的方法快了1.5倍。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Motion planning method based on safe trajectory tree network


    Weitere Titelangaben:

    一种基于安全轨迹树网络的运动规划方法


    Beteiligte:
    SUN HONGBIN (Autor:in) / CHEN LIMING (Autor:in) / CHEN WEIHUANG (Autor:in) / JI HAOXUAN (Autor:in) / YANG YANLONG (Autor:in) / ZHANG JINGMIN (Autor:in)

    Erscheinungsdatum :

    2023-11-21


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES / B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen




    Methodology for Safe Trajectory Planning

    Lazarowska, Agnieszka | Springer Verlag | 2022


    Agile and safe segmented trajectory planning method for spacecraft

    GUO ZIXI / TIAN KEFENG / GUAN XIN et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Tree Based Trajectory Planning for Mars Aerocapture

    Chakrabarty, Anjan / Swei, Sean Shan-Min / Prabhu, Dinesh K. | AIAA | 2016


    Safe trajectory planning for maritime surface ships

    Lazarowska, Agnieszka | TIBKAT | 2022