@article{QIAN2026, 
author = {Kun QIAN and Jiying DUAN and Chang QI and Baojun LI and Xueshu LIU and Haiyang LI and Mingheng ZHANG and Lei CAI},
title = {Design of an experimental method for evaluating ride comfort of intelligent driving vehicles},
year = {2026},
journal = {Experimental Technology and Management},
volume = {43},
number = {9},
pages = {40-49},
keywords = {intelligent driving, ride comfort, psychoacoustics, vehicle vibration, subjective evaluation},
url = {https://www.sciopen.com/article/10.16791/j.cnki.sjg.2026.09.006},
doi = {10.16791/j.cnki.sjg.2026.09.006},
abstract = {ObjectiveTo address the new challenges of ride comfort evaluation in intelligent driving vehicles, this study proposes an experimental method that integrates acoustic and vibration responses. Unlike conventional human-driven vehicles, intelligent driving vehicles exhibit distinct motion-control characteristics in scenarios such as adaptive car-following, lane changing, acceleration, deceleration, and turning, which may lead to different occupant comfort perceptions. Existing evaluation methods mainly focus on vibration-related physical indices and are not fully suited to intelligent driving conditions. Therefore, an experimental framework that combines objective acoustic and vibration indices with subjective comfort evaluation is needed.MethodsFour typical intelligent driving scenarios, including adaptive car-following, automatic lane changing, acceleration/deceleration, and turning, were designed as the test conditions. An intelligent electric vehicle was used as the test platform, and all experiments were conducted in automated driving mode under controlled conditions. In-cabin noise signals were collected using binaural microphones, and seat vibration signals were recorded using a tri-axial seat-pad accelerometer, with both signals synchronously acquired to ensure temporal consistency. Based on the measured data, psychoacoustic parameters and vibration indices were extracted to characterize ride comfort–related stimuli. The acoustic indices mainly included loudness, sharpness, and roughness, whereas the vibration indices included root-mean-square values and jerk-related parameters. Subjective comfort scoring was carried out immediately after each test scenario, and consistency analysis was conducted on the subjective data to ensure rating reliability. In addition, all objective indices were normalized, and Spearman correlation analysis was used to examine their relationships with the subjective comfort scores.ResultsResults show that the proposed method can effectively distinguish ride comfort characteristics across different intelligent driving scenarios. In terms of vibration response, acceleration/deceleration scenarios showed relatively strong longitudinal excitation, whereas lane changing and turning scenarios exhibited more pronounced lateral vibration characteristics. Adaptive car-following, by contrast, produced relatively smooth vibration changes. Regarding acoustic response, scenarios involving stronger control actions generally produced higher psychoacoustic parameter values, indicating more noticeable auditory stimulation. Subjective evaluation results demonstrated good consistency, confirming the reliability of the experimental design and scoring process. Correlation analysis further showed that the main objective indices were negatively correlated with subjective comfort scores: psychoacoustic parameters such as loudness, sharpness, and roughness exhibited relatively clear correlations with subjective ratings, while some vibration indices also reflected variations in comfort across scenarios. These results indicate that both acoustic and vibration factors play an important role in ride comfort evaluation for intelligent driving vehicles and that combining them can better characterize occupant perception.ConclusionsA ride comfort evaluation method for intelligent driving vehicles was established in this study, including scenario design, synchronized signal acquisition, subjective evaluation, objective feature extraction, and correlation analysis. The method effectively characterizes comfort differences among typical intelligent driving scenarios and reveals the relationship between objective acoustic–vibration responses and subjective perception. Compared with traditional evaluation approaches, the proposed method better captures the characteristics of intelligent driving conditions and provides a practical reference for experimental teaching, ride comfort testing, and the development of evaluation methods for intelligent vehicles.}
}