Car safety is of paramount importance to buyers

Integrating advanced driver assistance systems to improve driver safety

An integrated driver assistance concept for avoiding accidents and mitigating the consequences of accidents in urban traffic scenarios is at the centre of a collaborative project between Darmstadt Technical University and international automotive supplier Continental.

The integrated system links various advanced driver assistance systems (ADAS), enabling a high-performance system architecture with more powerful functions for perceiving vehicle surroundings than is possible using ADAS alone.

“To date, assistance systems such as Lane Keeping Assistant or collision warning have functioned as independent systems within the vehicle. By interlinking them, we can make optimum use of the existing sensor infrastructure in the vehicle,” said Professor Hermann Winner, Proreta 3 project manager at TU Darmstadt.

Proreta 3 is the third collaborative research project,and is now entering its last phase. During the first phase of the project, the jointly developed concept, based on a modular system architecture, was incorporated into a Continental research vehicle. This last phase of the project will focus on testing how the system works in practice by demonstrating the effectiveness of the concept under real-life driving conditions.

To monitor the area around the vehicle, the research vehicle uses a production-ready sensor system fitted with a stereo camera and a long-range radar sensor at the front and four short-range radar sensors for monitoring the area to the side and rear of the vehicle.

Based on this system of sensors, the researchers at TU Darmstadt created what they call a “safety corridor”. If the vehicle strays outside the safety corridor due to driver error or distraction, the driver is alerted and assisted where necessary through active system intervention with steering and braking.

“In complex urban traffic scenarios, the interaction between driver and assistance system must be optimised so that the driver understands the instructions intuitively and can respond accordingly,” said Dr Peter Rieth.

The research vehicle is kitted out with an interior infrared camera, identifying the direction in which the driver is looking, to help establish if the driver is looking at the road ahead or somewhere else. Further information is provided by the way the driver operates the steering wheel, accelerator and brake pedal. The driver assistance systems can then provide data on the level of risk.

The human-machine interface (HMI) features of the system include an intelligent accelerator pedal, which is able to communicate with the driver through exerting counterpressure or vibrating.

An LED strip light known as the Halo is linked to the interior infrared camera and the various driver assistance systems. This forms the core element of the HMI and is designed to direct the driver’s attention to hazards around the vehicle depending on the surroundings and driver status in cases where a hazard is detected outside the driver’s field of vision.

In such a case, an early warning can be issued and the driver can then respond accordingly. If the hazard is within the driver’s field of vision, then only an acute alert is issued. If the driver is visually distracted, a warning can be issued if the situation has not yet become critical.

The Halo covers the interiors of the car and can be illuminated in various colours to reflect different levels of urgency. The driver perceives this peripherally, drawing attention almost intuitively back in the right direction.

Within the safety corridor, partially automated driving is also possible with the aid of a longitudinal and lateral guidance system – the system can perform certain driving manoeuvres independently, thereby supporting and relieving stress on the driver.

The project will run until autumn 2014.

Recent articles

Info Message

Our sites use cookies to support some functionality, and to collect anonymous user data.

Learn more about IET cookies and how to control them