Vehicle Interior Occupant Monitoring

Research at THI in the field of OMS has progressed dramatically over the last 10 years, ranging from building small driver monitoring systems for cockpits to driver distraction systems and multimodal SOTA detection of occupants with the fusion of optical and radar systems to measure different parameters for adaptive airbag deployment.

With advances in automotive technology, ensuring safety and comfort in vehicles has become a primary concern. The increasing automation of vehicles allows occupants to take over non-driving tasks, which affects their seating position and favors so-called out-of-position situations that pose a major safety risk. In other words, occupants could be killed/injured even if safety systems are used. At best, a system is needed that takes such conditions into account and passes this information on to the safety systems before they react. This is where the vehicle interior monitoring system comes into play. These systems are either optical system bases or radar systems or a fusion of both systems with other sensors of the vehicle that measure the occupants' positions, poses, gestures/actions and vital signs and use this information for low-stream safety systems (airbags/safety belts) that help analyze the occupant's status during the crash or uncertainty and help estimate the severity of the crash, along with external perception data for adaptive or safe deployment of the airbag for a specific person. This system is also able to recognize the different occupant classes (large man, small woman, child or infant) and decides to adapt the deployment of the airbag according to FMVSS guidelines.

 

Completed Projects

No Deploy Zones I

Revolutionizing occupant safety by intelligently determining airbag deployment based on real-time detection of occupant positions.

No Deploy Zones II

Technology to precisely map occupants' head positions in autonomous vehicles, enhancing passenger safety by customizing airbag deployment and other safety features according to specific in-car scenarios.

InHealth

Innovative approach in automotive health technology to assess and enhance driver safety under various conditions like stress or sleepiness.

Camera based Distraction System

Evaluation of the effectiveness of system, focusing on accurately detecting driver inattention and drowsiness with limited processing requirements.

InSens

IFAS

Multi-modality based Occupant Monitoring System for Adaptive Activation of Restraint Systems

SAFE-UP

Development of the SOTA proof of concept of Vehicle Interior Occupant Monitoring System

Publications and Patents

2023
  • Palandurkar, T. V., Chan, L. Y., da Silva, J. L., Zimmer, A., & Schwarz, U. T. (2023, May). Driver’s chest position detection using FMCW radar data collected in a vehicle mock-up and CNN. In 2023 24th International Radar Symposium (IRS) (pp. 1-12). IEEE. (link)
2021
  • L. G. T. Ribas, M. P. Cocron, J. L. da Silva, A. Zimmer and T. Brandmeier, "In-Cabin vehicle synthetic data to test Deep Learning based human pose estimation models," 2021 IEEE Intelligent Vehicles Symposium (IV), Nagoya, Japan, 2021, pp. 610-615, doi: 10.1109/IV48863.2021.9576020. (link)
2014
  • J. L. da Silva, T. Brandmeier., A. Zimmer. Automatic Measurement of Driver Attention Level via Computational Vision. Brazilian Congress of Biomedical Engineering (CBEB). 2014. (link)

Contact

Guest professor CARISSMA
Prof. Dr. Alessandro Zimmer
Phone: +49 841 9348-6404
Room: K206
E-Mail: