This study evaluates the risk level of human-machine collaborative driving takeover in a highway environment under the interaction of non-driving related tasks and takeover request prompt scenarios. Using a driving simulator, a 5x5 factor analysis examined non-driving tasks and takeover prompts. Takeover impact and risk indicator features were extracted, with risk indicators weighted using an improved ANP method. On this basis, the XGBoost algorithm was employed to identify crucial variables that reflect the level of takeover risk and to construct a driver takeover risk assessment model. The evaluation results indicate that the risk indicator feature with the greatest influence on the takeover risk level was the minimum TTC, which had the highest correlation (r=-0.81); the scene factor in the takeover influence feature had the highest correlation with the takeover risk level (r=-0.78), which had the greatest influence on driver takeover safety; Although non-driving related tasks exhibited a weak correlation with takeover reaction time, steering reaction time, and minimum TTC, the effect was minor. The XGBoost algorithm-based risk assessment model demonstrated superior performance over LightGBM and SVM, with 87.1% accuracy. Overall, this study highlights the significant influence of takeover scenes and minimum TTC on collaborative driving risk, enabling accurate risk modeling.

This paper investigates the effects of non-driving related tasks, take-over request time, and take-over mode interactions on take-over performance in human–machine cooperative driving in a highway environment. Based on the driving simulation platform, a human–machine collaborative driving simulation experiment was designed with various take-over quality influencing factors. The non-driving related tasks included no task, listening to the radio, watching videos, playing games, and listening to the radio and playing games; the take-over request time was set to 6, 5, 4, and 3 s, and the take-over methods include passive and active take-over. Take-over test data were collected from 65 drivers. The results showed that different take-over request times had significant effects on driver take-over performance and vehicle take-over steady state (p < 0.05). Driver reaction time and minimum TTC decreased with decreasing take-over request time, maximum synthetic acceleration increased with decreasing take-over request time, accident rate increased significantly at 3 s take-over request time, and take-over safety was basically ensured at 4 s request time. Different non-driving related tasks have a significant effect on driver take-over performance (p < 0.05). Compared with no task, non-driving related tasks significantly increase driver reaction time, but they only have a small effect on vehicle take-over steady state. Vehicle take-over mode has a significant effect on human–machine cooperative driving take-over quality; compared with passive take-over mode, the take-over quality under active take-over mode is significantly lower.