Frequently Asked Questions
There is no conventional traffic model behind it at all. The traffic is its own model. We work around the data.
The most expensive and severe errors occur usually right at the beginning of the development in the phase of specification and conception. Therefore it is of extremely high relevance in the development and implementation of complex systems to identify conflicts and unrealistic requirements as soon as possible so as to avoid and/or eliminate potential errors in the concept.
Not at all! The issue of car-2-car- and car-2-infrastructure-communication makes the application of sophisticated traffic control algorithms even more important, because these have to become much more functional and intelligent then. Especially in the case of saturated traffic flows with high traffic volumes individual cars will not be able to sort out themselves without the introduction of superordinate traffic control instances for a fair control regime to balance the individual interests.
Only because the traffic control units are not necessarily physically linked to traditional traffic light systems anymore and these are quasi virtualized, this does not mean that traffic control algorithms can be omitted!
VERONET exactly addresses this fact already yet by its subsidiary hierarchical control architecture.
The implementation requirements of VERONET can be found here.
VERONET does not need a specific traffic sensor technology and can deal with any arbitrary available sensor data and information for the evaluation of traffic. Either Floating Car Data or stationary traffic sensors can easily be integrated and combined, as long as the data can be read in realtime. The effectively required traffic sensors are determined by the different traffic situations, which have to be handled according the results of the requirements management. There are also methods and procedures available, which help to identify and specify the necessary sensors more or less automatically.
We apologize for being too abstract sometimes. However, the criticised abstractness is absolutely necessary to achieve standardization of the uniform control objects and the underlying automation. Finding the right level of abstraction is the main work and art of each successful software engineer, mathematician, or indeed of every scientist. Unfortunately the right level of abstraction is not due to the beholder but due to the given complexity of the problem. And the complexity is not insignificant here. Nevertheless, enough examples should be available now to illustrate the solutions more intuitively or to conceal the abstractness in the available software solutions.
Perhaps also the following articles may help for clarification:
- Traffic Technology International, Juli 2012: Anomalie and change detection for intelligent traffic control
- Traffic Technology International, August/September 2012: Intelligent algorithms make the difference
- Traffic Technology International, April/Mai 2013: Big Data - what then?
- Traffic Technology International, Juli 2013: Traffic control software for improved adaptability
- Traffic Technology International, Februar/März 2014: Game-like control interfaces for signal operations
- Traffic Technology International, August/September 2015: A self-learning system optimizes active traffic management
- Traffic Technology International, August/September 2016: Will autonomous driving solve our traffic problems?
- Traffic Technology International, August/September 2017: Balacing the benefits of automated driving
The introduction of automated and connected driving opens the opportunity, that traffic control can be reinvented or redefined for the realization of real improvements in traffic efficiency.
VERONET provides the technological framework and underlying architecture, if automated driving should be used for real improvement in traffic efficiency and traffic safety in common.
The approach that has been developed is inspired by natural solutions in keeping with bionics. Like in an ant colony or bee hive there is no central intelligence. The ant colony consists of numerous simple individuals, which only follow simple and restricted tasks with a set of only a few simple rules. Nevertheless, the ant colony builds up a virtual creature with astonishing intelligence by the simple but neat interaction of the ants, which are individually not informed about the overall situation. That way VERONET is a kind of swarm intelligence.