Personalised lane keeping assist strategy: adaptation to driving style

Rath, Jagat Jyoti; Senouth, Chouki; Popieul, Jean Christophe · 2019 · Crossref

DOI: 10.1049/iet-cta.2018.5941

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the need for personalized Advanced Driver Assistance Systems (ADAS) that adapt to individual driving styles (DS) to improve human-machine interaction and lane keeping performance. While existing ADAS architectures often treat drivers uniformly, this work proposes a personalized lane keeping assist (PLKA) strategy that modulates assistance based on the driver’s specific style, categorized by aggressiveness. The motivation stems from the observation that driving style significantly influences lateral vehicle dynamics and driver comfort, particularly during navigation of high- and low-curvature tracks. The methodology involves a co-operative control architecture integrating a non-linear vehicle model, a visual-cues-based human driver model, and a robust controller. First, driving style is identified using a fuzzy logic classifier based on statistical analysis of lateral jerk and steer feel signals. These signals are normalized relative to driving zones defined by lateral acceleration and road curvature. The classifier categorizes drivers into four styles: calm, moderate, aggressive, and very aggressive. Second, an adaptive activity function (AAF) and an assistance modulation factor (AMF) are formulated to determine the required level of assistance based on the identified style and driver torque. Finally, a robust higher-order sliding mode (HOSM) controller generates the assistive torque to minimize lane deviation errors (lateral deviation and heading error) while ensuring driver comfort. Closed-loop stability of the driver-vehicle system is established in the presence of disturbances. The proposed architecture was validated through numerical simulations on the Satory test track, which features road sections with radii varying from 25 to 500 meters. The driver model used in the simulation was validated against real driver data from the SHERPA simulator. Results demonstrated that the fuzzy classifier effectively identified driving styles based on lateral jerk and steer feel. The co-operative control system successfully adapted the assistive torque according to the driver’s style, maintaining lane keeping performance across different curvatures. The HOSM controller ensured robustness against disturbances, and the modulation function prevented excessive assistance that could cause discomfort, thereby achieving a balance between safety and driver comfort. The significance of this work lies in its contribution to personalized ADAS design, specifically for lateral motion control. By adapting assistance to driving style, the system enhances human-machine interaction and overall efficacy. The study provides a framework for integrating driver behavior recognition with robust control strategies, offering a pathway for more intuitive and comfortable semi-autonomous driving systems. The approach highlights the importance of considering dynamic driver attributes in the design of co-operative control architectures for future intelligent vehicles.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• lane positioning
• steering pattern

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: behavioral performance data
• Methodological Resource: tool software
• Theoretical Contribution: computational model