Autonomous Driving: ÖV-LeitmotiF-KI

To achieve these ambitious objectives, seven experienced project partners from industry and science are jointly researching the standard for integrating automated vehicles into control systems. In addition to INIT, these include renowned scientific institutions such as the Karlsruhe FZI Research Center for Information Technology, the Karlsruhe Institute of Technology (KIT) and the Ilmenau University of Technology (TUIL). On the industry side, ESE Engineering und Softwareentwicklung GmbH (engineering and software development) and IVU Traffic Technologies are involved in the project, which is being coordinated by the VDV.

Together with IVU and the FZI, INIT is analysing manufacturer-independent interfaces as part of the project. Specifically, the aim is to define the structure of agnostic interfaces for seamless integration of control systems, vehicle components (on-board computers), and AD-systems (AD = automated driving) from any manufacturer of autonomous vehicles. To this end, interfaces based on VDV document 435 were specified during the project and implemented in the IVU and INIT systems for an upcoming field test with two autonomous shuttles from FZI. The aim is to test dispatching measures as well as normal operation, also with regard to the interoperability of systems from different manufacturers.

A reference architecture was created as the basis for the specifications and implementation. Components and interface descriptions were also defined. This reference architecture can be divided into control centre-side and vehicle-side components as well as PT- and AD-side components and responsibilities. In addition to standard PT-PT interfaces, e.g. for timetable data transmission, there are also PT-AD interfaces, e.g. for mission transmission. Missions are commands that are created from timetables and routes (from the public transport side) to describe how the vehicle is to travel from one point to another. This may also include the information that passengers will board/alight at the end of a mission. The AD side then generates more precise information from these missions for the vehicle so that it can drive autonomously.

The separation into PT- and AD-side components also has safety-related aspects: The necessary certifications of the safety-critical components are covered by the AD systems, so that the public transport side remains flexible while retaining sovereignty over the public transport functions. The Ilmenau University is using a laboratory set-up – a replica of a vehicle interior – to shed light on interaction with passengers in autonomous vehicles. For example: Which and how many buttons (one or more) must there be in the vehicle so that passengers can call up information about the journey, but also call for help in emergency situations? In short: What is needed in the vehicle interior to replace drivers? 

In addition, the content developed so far in the project for a new standard was evaluated from the user’s perspective, e.g. dispatchers in public transport companies, as they are the ones who will be working with it in the future. In the project, ESE provides a safety-related analysis of the developed project content in order to fulfil the legal requirements. The FZI is providing shuttle buses from the company EasyMile with a self-developed autonomous driving function. The planning of the field test on the East Campus of the KIT is also being carried out by the FZI in close co-operation with IVU and INIT. KIT is playing a leading role in the development of the reference model and also contributing its expertise in the field of AI to the project.

The development of a new VDV standard for the integration of autonomous vehicles into existing public transport systemsshould make the integration process easier for public transport companies in the future and promote interoperability between different systems. By integrating AD vehicles into the public transport systems as seamlessly as possible, these vehicles can be treated in a similar way to vehicles conventionally operated by drivers, e.g. by carrying out a dispatching measure.

The importance of autonomous vehicles in the future is beyond question. INIT therefore also took the research project asan opportunity to incorporate automated vehicles in the further development of the MOBILE-ITCS nextGen (Intermodal Transport Control System) in order to enable public transport companies to monitor autonomous and nonautonomous vehicles in one operations control system in the future.