The invention is suitable for the technical field of automatic driving, provides an automatic driving map enhanced trajectory prediction method integrating semantic information, breaks through the limitation that a traditional trajectory prediction model only considers a vehicle state, and jointly inputs map information, a vehicle historical trajectory, a speed and an acceleration into a neural network. By means of stable and accurate map data, a more perfect context background is provided for prediction, and the limitation of sensor data is made up. And meanwhile, an advanced Transform encoder-decoder architecture is used for predicting the vehicle track, so that the adaptability of multi-modal data of the algorithm is improved. The limitation that only a single data source is used in a traditional method is broken through, the system can better predict diversified driving behaviors, and higher adaptability to multi-modal data is shown. In the model testing process, the performance of the automatic driving map enhanced trajectory prediction method based on the comprehensive semantic information is verified, the average displacement error can be reduced by 33% at most, and great advantages are achieved.

    本发明适用于自动驾驶技术领域,提供了一种综合语义信息的自动驾驶地图增强轨迹预测方法,突破了传统轨迹预测模型仅考虑车辆状态的限制,通过将地图信息与车辆历史轨迹、速度、加速度联合输入到神经网络。借助稳定、精确的地图数据,为预测提供更完善的上下文背景,弥补传感器数据的局限性。同时使用先进的Transformer编码器‑解码器架构对车辆轨迹进行预测,提升了算法多模态数据的适应性。突破了传统方法仅利用单一的数据源的限制,使系统能够更好地预测多样性的驾驶行为,对多模态的数据表现出更高的适应性。在模型测试过程中验证了本发明提出的综合语义信息的自动驾驶地图增强轨迹预测方法的性能,其平均位移误差最多可降低33%,具有很大的优势。


    Access

    Download


    Export, share and cite



    Title :

    Autonomous driving map enhanced trajectory prediction method integrating semantic information


    Additional title:

    一种综合语义信息的自动驾驶地图增强轨迹预测方法


    Contributors:
    ZHAO RUI (author) / QIAO SENYAO (author) / WU JIANGHANG (author) / LI HAOCHENG (author) / YAN JIAQI (author) / GAO FEI (author) / GAO ZHENHAI (author)

    Publication date :

    2024-08-09


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06V / 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 / G01C Messen von Entfernungen, Höhen, Neigungen oder Richtungen , MEASURING DISTANCES, LEVELS OR BEARINGS / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Advancing Autonomous Driving Safety Through LLM Enhanced Trajectory Prediction

    Cheng, Qian / Jiao, Xinyu / Yang, Mengmeng et al. | Springer Verlag | 2024

    Free access

    Intention-Driven Trajectory Prediction for Autonomous Driving

    Fan, Shiwei / Li, Xiangxu / Li, Fei | IEEE | 2021



    GRIP++: Enhanced Graph-based Interaction-aware Trajectory Prediction for Autonomous Driving

    Li, Xin / Ying, Xiaowen / Chuah, Mooi Choo | ArXiv | 2019

    Free access

    INTENTION-DRIVEN TRAJECTORY PREDICTION FOR AUTONOMOUS DRIVING

    Fan, Shiwei / Li, Xiangxu / Li, Fei | British Library Conference Proceedings | 2021