BUP20 - TEDZ

Autonomous driving enables new seating positions that pose safety risks that are not immediately apparent. Occupant monitoring systems to monitor posture, including feedback, are a promising safety measure to minimize the risk of injury. However, they are often expensive and poorly available.

In this project, a smartphone-based monitoring system is being developed that not only monitors occupant posture, but simultaneously performs a safety assessment based on highly complex simulation models and provides feedback in critical cases. This allows occupants to enjoy the benefits of autonomous driving without exposing themselves to new risks.

Aim and approach

The project plans to:

• derive efficient human models based on the predicted injuries of advanced crash simulations, which can assign a risk of injury to the body posture of a passenger and are real-time capable through a combination of model reduction and machine learning    
• reconstruct the occupant's positions and posture using camera footage from a smartphone
• Implement and compute a transportable real-time capable digital twin of the occupant on the smartphone.

The implementation of the transportable digital twin with visualization of the posture on the smartphone. Safe posture (left). Critical posture (right) with crushing of internal organs/damage to knee joints.