Thanks to the automatic cooperation between different road users, totally new applications can be developed that contribute to increased road safety and efficiency. One example of this is lane closures, where cooperative lane merging assistants can reduce congestion.
As a result of the innovative nature of these kinds of functions based on intensive networking, they are often totally new and customised developments. The first functions implemented work with simple interaction mechanisms like the unidirectional triggering of a warning with a congestion warning assistant and can be produced using current software tools. Future cooperative driving functions on the other hand have complicated interaction mechanisms, for example as required for automated lane merging. They involve different players, for example other vehicles or back end services, which contribute crucially to the outcome of driving decisions.
Design and Modelling
Current methods and tools do not offer an approach that takes functions across all models into account at all. For this reason, ESK researchers are optimizing and developing modelling methods to include and cover the requirements of interactively networked vehicles in the future at an early stage. At the same time, this modelling can also be used as the basis for further development and when prototyping cooperative functions.
Simulation for Parameterization
Cooperative driving functions cannot be tested on the roads in the early stages of development for safety and cost reasons. Developers use a wide range of simulation tools instead. Starting with the extensive modelling of the system, the required components and even the environment, modules and configurations can be automatically produced for the respective tools using the ESK environment. A combination of tried, tested and validated simulators from the transport, driving dynamics and communication sectors allows for the planned functions to be analysed with a high degree of detail and continuously improved. Statements about the system’s behaviour when interacting with several vehicles, about the dimensioning and parametrising of algorithms and possible boundary conditions and limits can be really easily determined and considered in the ongoing process.
Real and Virtual Prototyping
Once the function has proved itself in simulation, it can be directly transferred from the model to a prototype with little effort and cost and then tested in a vehicle in real conditions. For this, the solution that is being worked on at ESK provides an execution environment where the same code basis is used as in the simulation. In this way implementing the same function several times in different environments can be avoided, which reduces errors and enables seamless switching between the different phases of development. A great deal of data is collected during the course of testing in real vehicles. The measurement and collection points required for this are already defined in the model. As a result it is possible to integrate the results from the test drives directly into the model again and analyse them using standardized tools without further ado.
Verified and Safe Model
The result of this process is a verified and safe model of a driving function. This model can be transferred automatically into a specification for a supplier or implemented in a series vehicle.
Fraunhofer ESK provides a tool chain for the integrated development, evaluation and prototyping of distributed cooperative driving functions. As a result new findings from test drives are integrated into the model and evaluated first in the simulation environment before elaborate test drives have to be carried out again. In addition, developers of these kinds of functions can use model-based methods and tools from the model-based design to verification and simulation to prototyping.
Vehicle manufacturers, suppliers and manufacturers of transport infrastructures can also continue to use their own tools. Fraunhofer ESK can help here with the analysis of their own tool chains and by closing any gaps in integrated development through optimization.