Opel demonstrates highly Automated Driving as a part of the research project Ko-HAF

The automaker is part of Ko-HAF "“ "Kooperatives hochautomatisiertes Fahren" "“ a German project researching cooperative highly automated driving, which began in June 2015. The Opel Insignia prototype displayed the functions of cooperative highly automated driving during the presentation of Ko-HAF's final results at the Opel Test Center Rodgau-Dudenhofen.

Cooperative highly automated driving systems do not need to be supervised by the driver all the time. Drivers can perform other tasks, but when prompted by the system they must be able to take control of the vehicle within a certain time period. The vehicle must therefore be able to "see" further ahead than possible with its own sensors.

In Ko-HAF, vehicles send information about their current road environment, such as construction sites, traffic jams and accidents, to a Safety Server. The information is collected and processed by the Safety Server, so that a precise map is available to vehicles when they request it "“ like an artificial horizon that delivers a highly detailed preview of the road ahead.

Opel's role in Ko-HAF focused on the computerised maps and the process of disengaging the car from the automated driving condition, thus returning control to the driver. The engineers from Ruesselsheim designed the architecture, interfaces and protocols of the Safety Server, which were evaluated in the project.

An additional core-task was the development of a self-localisation method for the vehicle. Opel designed algorithms for visual mapping and localization which are merged with information from back-end and onboard maps, onboard sensors and the Global Navigation Satellite System (GNSS). The localisation method was validated on the Opel Insignia test car at the Opel Test Centre and on the Ko-HAF test route on the motorways around Frankfurt am Main.

Opel's second area of focus concerned the driver's actions, so-called non-driving tasks. The company's engineers developed software and a system of sensors to detect and classify the driver's actions while the car is driving automatically.