Evaluation method for driver comfort under multi axis coherent vibration of seats

Area of Science:

• Automotive Engineering
• Human Factors Engineering
• Biomedical Engineering

Background:

• Ergonomics research highlights the importance of seat design for driver comfort and health.
• Understanding the impact of multi-axis coherent vibration on drivers is crucial for vehicle development.

Purpose of the Study:

• To evaluate the impact of seat multi-axis coherent vibration on driver comfort using deep learning.
• To develop a quantitative method for assessing driver comfort based on vibration data.

Main Methods:

• Collected multi-axis vibration signals from vehicle seats (backrest, cushion, floor) during road tests.
• Gathered simultaneous subjective driver comfort evaluations.
• Validated subjective-objective data consistency using Stevens' power law (R² > 70%).
• Developed and applied a deep learning model integrating multimodal coherent features for quantitative comfort assessment.

Main Results:

• Subjective driver comfort evaluations were found to correlate well with objective vibration data.
• The deep learning model accurately captured comfort-affecting frequency characteristics.
• The model achieved high performance metrics (R²=0.931, RMSE=0.096, MAE=0.071), comparable to expert evaluations.

Conclusions:

• The proposed deep learning method offers a reliable and accurate approach for quantitative driver comfort evaluation.
• This method has significant implications for improving driver health, comfort, and safety through optimized seat design.
• The findings support the integration of advanced AI techniques in automotive ergonomics research.