A consortium of automobile manufacturers, automotive suppliers and public partners has launched in Germany the Ko-HAF (Cooperative Highly Automated Driving) research project. The aim of the project is to research new systems and functions permitting highly automated driving at higher speeds and in more complex situations.
A consortium of automobile manufacturers, automotive suppliers and public partners has launched in Germany the Ko-HAF (Ko-HAF – Cooperative Highly Automated Driving) research project. Ko-HAF has a total budget of 36.3 million euros and is expected to run until November 2018. It is being supported by Germany’s Federal Ministry for Economic Affairs and Energy (BMWi) as the first project within the framework of its new program “New Vehicle and System Technologies”. The international automotive supplier Continental is taking on coordination of the Ko-HAF project. Cooperative refers to the interaction between several highly automated vehicles.
The aim of the project is to research new systems and functions permitting highly automated driving at higher speeds and in more complex situations. The project will develop standards and technologies that enable cooperative driving between highly-automated vehicles in everyday road traffic – for example merging with traffic on highways. The project work covers various subject areas from computer simulations to test drives on closed sites and later on public roads.
During highly automated driving, the driver no longer has to monitor the systems permanently and can instead leave the chore of driving to the vehicle for a certain time. However, the driver has to be able to take over the driving task within a certain time. For the time that the driver needs to do so, it is necessary for the vehicle to correctly assess its surroundings and the traffic situation. This is a particular challenge at higher speeds and in more complex scenarios. This is one of the tasks that Ko-HAF is looking into: the starting point is a backend solution in which the vehicles communicate with one another via a server, aided by mobile radio (LTE/UMTS). The server collates and evaluates information about the vehicle surroundings, and makes it available to the vehicles in a consistent form. This method of forward-looking driving is necessary for high automation in more complex situations.
The backend solution is based on vehicles communicating wirelessly with a central server. For the duration of the project, a variety of vehicles belonging to the different project partners will collect the data generated by their on-board sensors during test drives and feed the server with information about their surroundings. Along with information on objects in the road, this also includes data on the existence and quality of lane markings. The collected information will be analyzed and consolidated on the server. An expanded digital map will then be made available for the vehicles to download. This will enable them to be informed in plenty of time about traffic hazards behind a bend or just over the crest of a hill.
In addition to the backend solution, the Ko-HAF project is also working on precise self-localization for highly automated vehicles. For instance, the vehicles must be able to identify which lane they are traveling in as well as their exact position in the lane. One of the project’s other focal points is how to integrate drivers into the process. Other areas being explored by the consortium include how a highly automated system can be made secure and tested.