Advanced Driver Assistance Systems

Advanced Driver Assistance Systems

from: http://www.rcs.ei.tum.de/en/research/driver-assistance/

Advanced Driver Assistance Systems (ADAS) are helping drivers to operate vehicles in a conveniant and safe way. While current systems mostly focus on a warning, some systems in the market are already initiating evasive actions. In research, tele-operated or even autonomous driving is no longer visionary.

This trends lead to new challeges in the integration with existing systems and for future developments, e.g.:

• Integration of a growing numer of ADAS funcionalities per vehicle
• Sensors are driving several applications with heterogeneous requirements
• Sensor information gets shared among vehicles
• Applications evolve from reactive systems to predictive and cognitive systems.
• Highly specialized applications for adverse environmental conditions
• Electric Vehicles have to cope with limited energy buffers

Many of our research areas are also highly relevant for the specific requirements in the ADAS domain, e.g.

• Control/Architecture Co-Design
• ECU consolidation
• Admission Control

Computer Vision for ADAS

Furthermore on an application level, we focus on vision-based ADAS. So far, many ADAS incorporating cameras heavily rely on a data fusion with other sensor information gathered by mostly active (= emmiting) sensory like radar and lidar. The latter are a major cost factor. Improving vision systems helps  reducing overall vehicle costs. Human drivers mostly rely on the visual channel while operating a vehicle - improvements in image understanding are expected to lead to a sigificant better performace of ADAS.

Outdoor camera systems have to cope with a highly unstructured and changing environent (e.g. variety of objects and traffic infrastructure, intrinsic ambiguities, whether, daytime), which limits the range of application. While whether and daytime changes, the specific requirements for vision tasks change as well. One of our challenges is to manage and utilize sunligth effects and direct cast vehicle shadows for automotive vision systems.

Contact person:

Florian Rattei