Abstract
This study aims to create a system that can evaluate railway vehicle characteristics while simultaneously controlling human body behavior using numerical simulations. The proposed system consists of a vehicle model, a human motion generation model, and a musculoskeletal model. The vehicle model first calculates the vehicle motion in various situations and then calculates the human motion by imputing the acceleration data of the vehicle seat to the human model. At this point, the human motion is calculated by inverse and forward dynamics with a reaction force at the contact point between the human body and the seat. Finally, muscle activation and loading are calculated by inverse dynamics with the human body position, and the reaction force is determined from the above calculations. Using these data, it is possible to examine and evaluate the human body posture. Note that this paper limits discussion to the relationship between the human body motion generation and the railway vehicle. The muscle force will be left for a future study. To demonstrate the effectiveness of the human motion generation model, a simple human dummy model that consists of only a spring and damper was created. Comparing this dummy model (which simulates the experimental model used for experiments such as automobile collisions) with the human motion generation model, it can be concluded that the human motion generation model can simulate details of human motion that the simple dummy model cannot. Thus, the control system in the human motion generation model is important for human motion evaluations.
