Object detection is a critical component of a self-driving system, tasked with inferring the current states of the surrounding traffic actors. While there exist a number of studies on the problem of inferring the position and shape of vehicle actors, understanding actors' orientation remains a challenge for existing state-of-the-art detectors. Orientation is an important property for downstream modules of an autonomous system, particularly relevant for motion prediction of stationary or reversing actors where current approaches struggle. We focus on this task and present a general method that allows us to more accurately infer vehicle orientations, which we apply to the state-of-the-art models that perform joint object detection and motion prediction. In addition, the approach is able to quantify prediction uncertainty, outputting the probability that the inferred orientation is flipped, which allows for improved motion prediction and safer autonomous operations. Empirical results show the benefits of the approach, obtaining state-of-the-art performance on the open-sourced nuScenes data set.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Uncertainty-Aware Estimation of Vehicle Orientation for Self-Driving Applications


    Contributors:


    Publication date :

    2021-09-19


    Size :

    1841023 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Long-Tail Prediction Uncertainty Aware Trajectory Planning for Self-driving Vehicles

    Zhou, Weitao / Cao, Zhong / Xu, Yunkang et al. | IEEE | 2022


    Uncertainty-Aware Behavior Planning in Automated Driving

    Bey, Henrik | TIBKAT | 2024

    Free access

    Position and orientation estimation method for self-driving car

    CHEN SHENGJUN / XIA HENG / XIAO ZHIGUANG et al. | European Patent Office | 2015

    Free access


    Vehicle orientation estimation system and vehicle orientation estimation method

    YU ZIHAN / YASUKAWA TAKAKIYO / IRIE KOTA et al. | European Patent Office | 2025

    Free access