Identifying edge cases for designed algorithms is critical for functional safety in autonomous driving deployment. In order to find the feasible boundary of designed algorithms, simulations are heavily used. However, simulations for autonomous driving validation are expensive due to the requirement of visual rendering, physical simulation, and AI agents. In this case, common sampling techniques, such as Monte Carlo Sampling, become computationally expensive due to their sample inefficiency. To improve sample efficiency and minimize the number of simulations, we propose a tailored active learning approach combining the Support Vector Machine (SVM) and the Gaussian Process Regressor (GPR). The SVM learns the feasible boundary iteratively with a new sampling point via active learning. Active Learning is achieved by using the information of the decision boundary of the current SVM and the uncertainty metric calculated by the GPR. The optimal sampling point is selected through multi-objective optimization so that the sampling point is close to the decision boundary of the current SVM as well as it has a high uncertainty metric. To illustrate the characteristics and effectiveness of the proposed approach, we apply the proposed approach to both simple problems with synthetic feasible regions and a complex industrial problem: autonomous driving path planning algorithm validation. With the proposed approach, edge cases are more effectively identified, so the classification accuracy of the resulting SVM is considerably higher than that of the SVM trained with conventional sampling techniques such as Monte Carlo Sampling and Latin Hypercube Sampling.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Active Learning Optimization for Boundary Identification Using Machine Learning-Assisted Method


    Additional title:

    Sae Int. J. Adv. and Curr. Prac. in Mobility


    Contributors:
    Ryu, Joonjae (author) / Wei, Yifan (author)

    Conference:

    WCX SAE World Congress Experience ; 2022



    Publication date :

    2022-03-29


    Size :

    6 pages




    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English




    Machine-Learning-Assisted Optimization of Aircraft Trajectories Under Realistic Constraints

    Xu, Yifan / Wandelt, Sebastian / Sun, Xiaoqian et al. | AIAA | 2023


    Lunar Campaign Optimization Using Machine Learning

    Bartkiewicz, Jacob / Haws, Terry D. / Fuller, Michael E. | IEEE | 2022


    Lunar Campaign Optimization Using Machine Learning

    Bartkiewicz, Jacob P. / Haws, Terry / Fuller, Michael | AIAA | 2021



    Hull Form Design Optimization using ROM with Machine Learning and Active Subspace Methods

    Hwang, Wontae / Kim, Jewon / Choi, Seongim et al. | AIAA | 2023