Title: Research on modelling and extracting methods for personalised driver longitudinal operation features

Authors: Zhilin Jin; Juntao Kong; Chao Yang; Shilong Cao

Addresses: Department of Vehicle Engineering, Nanjing University of Aeronautics and Astronautics, 29 Yudao Street, Qinhuai District, Nanjing, 210016, China ' Department of Vehicle Engineering, Nanjing University of Aeronautics and Astronautics, 29 Yudao Street, Qinhuai District, Nanjing, 210016, China ' Department of Vehicle Engineering, Nanjing University of Aeronautics and Astronautics, 29 Yudao Street, Qinhuai District, Nanjing, 210016, China ' Department of Vehicle Engineering, Nanjing University of Aeronautics and Astronautics, 29 Yudao Street, Qinhuai District, Nanjing, 210016, China

Abstract: The characteristic parameters of driver manipulation are crucial for research into human-machine co-driving control strategies in intelligent vehicles. This paper proposes an experimental data-driven modelling and extraction method for accurately obtaining the longitudinal operation feature parameters of drivers. First, a driver manipulation experimental platform and a fuzzy data collection method are created in order to collect longitudinal manipulation experimental data for aggressive and non-aggressive drivers. The feature model is then built using an experimental data-driven deep neural network training method. Combining the feature model and PD theory, feature parameters are extracted using a hierarchical method. Furthermore, experiments and simulations are carried out for two types of drivers. The results demonstrate that the validation set of the feature model has an RMS error of less than 5%. The extracted feature parameters' values are within the longitudinal operation feature value range and can reflect the personalised driver longitudinal operation characteristics.

Keywords: personalised driver; longitudinal operation feature model; fuzzy data-collection method; deep neural network training method; hierarchical extraction method.

DOI: 10.1504/IJVD.2025.147509

International Journal of Vehicle Design, 2025 Vol.97 No.1, pp.46 - 73

Received: 14 Jul 2024
Accepted: 10 Jan 2025
Published online: 18 Jul 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article