Driving on the Autobahn: Opel will build a prototype for highly automated driving
Rüsselsheim. Any driver who spends a lot of time in the car often wonders if this time could be used more efficiently, for example, by reading the newspaper, doing emails, or simply enjoying the ride. Cars that drive automatically, which do not require permanent control from the driver, would make such dreams reality.
Among the major challenges to automated driving are the highly detailed, computerized maps (so that the car knows exactly about the road it is driving down) and the process of disengaging the car from the automated driving condition, thus returning control to the driver.
These are two areas of automated driving being studied intensively by Opel in “Ko-HAF – Kooperatives hochautomatisiertes Fahren”, a German project researching cooperative highly automated driving.
Experts estimate that it takes about ten seconds for a human being to turn his attention from another activity to driving. A car that drives automatically must therefore be able to recognize the driving environment and assess the traffic situation within that time-period.
In order to make an accurate assessment, the car also needs significantly more information that it can obtain from its sensors and the map of the onboard navigation system.
This is where “Ko-HAF” and Opel come in with their research into a back-end solution called the Safety Server. In “Ko-HAF”, cars communicate with the Safety Server, supplying data such as presence and quality of lane markings or objects on the road from their own onboard sensors. The Safety Server processes this information and then provides the cars with a forward-looking digital map, which the cars need for highly automated driving.
Opel is working especially on the communication between server and car – e.g. designing the server-architecture, defining the interfaces and implementing the communication protocols.
Opel’s second area of focus concerns the driver’s actions. The company’s engineers are developing software and a system of sensors to detect and categorize the driver’s actions while the car is driving automatically.
In order to predict how the driver might react when he or she must retake control, the correlation between driving situations and the driver’s actions inside the car is of special interest.
An additional core-task is the development of a self-localization process for the vehicle. Opel is designing algorithms for visual mapping and localization which are then merged with information from back-end and onboard maps, movement sensors and the Global Navigation Satellite System (GNSS).
Opel will build a prototype that can display the essential functions of cooperative highly automated driving on the Autobahn, i.e. automatically entering the Autobahn and merging with the traffic, driving on the Autobahn including over-taking maneuvers, and finally exiting the Autobahn automatically.
Initially Opel will test and validate the prototype hardware and the highly automated driving functionality on a proving ground. Testing will continue on public roads as soon as the prototype technology has reached a sufficiently advanced stage.
The “Ko-HAF” research initiative began in June 2015 and will run until November 2018. The project has a budget totaling 36.3 million euros and is supported by the Federal Ministry for Economy and Energy.
