Introduction

Nowadays, the various operation activities of publicroads is inconceivable without the use of electronic road side sensors and other devices. Various parameters are measured by on-line sensors along the network, which provide useful information about the road and the actual traffict hat serves as the basis for operational or traffic management measures. In the last two decades, the quality and the accuracy of road side sensors, and the data they provide have improved a lot. Conventional electro-magnetic vehicle sensors (e.g., induction loop detectors) are increasingly being replaced by cameras, radars, and LiDARs that can be used more widely. By using an analysis algorithm smart cameras can also detect a number of traffic disturbances, such as slow traffic, stop-and-go traffic, hard shoulder usage, wrong way driving, and obstacle on road. With the help of an infrared reflector, these cameras are able to detect traffic anomalies even at night. Thermal cameras can also beused for traffic monitoring, which are mainly used around tunnels, but can also be used in many other areas, such as pedestrian detection. Radars has traditionally been used to measure distance and, in connection with this, to determine the speed of vehicles. LiDARs can generate a point cloud of the environment to detect static and moving objects. As the amount of information grows, raising the level of automation, use of machine learning, and Artificial Intelligence are essential in the field of traffic management. Due to the increasing amount of more and more detailed information available from different sources and the huge amount of raw and processed data from modern sensors, we can get a much more detailed picture of the traffic situation. This also means that we can create advanced models that better replicate the real situation, and processes and can better predict the state of traffic on a network over a period of up to 30-60 minutes. This leads towards safer, more comfortable and more predictable transport for all. Magyar Közút (Hungarian Public Roads) is devoted in piloting, and implementing cutting edge technologies, and fostering researches in the field of Intelligent Transport Systems. As a member of the consortium researching the swarm behaviour of nearby cars (objects, or entities) in the framework of the RAJ project, Magyar Közút provides the infrastructure for data collection.

Intelligent data collection and analysis

By extending classical traffic data collection and analysis, collecting information about the group behaviour of road users can enhance these realistic models. The so called swarm behaviour-based approach is a promising methodology, it can help in understanding heterogeneous transport situations. For the purpose of RAJ project, an urban traffic light signal is edjunction was equipped with fixed traffic monitoring cameras, covering the whole junction area.

By using this traffic monitoring scheme, machine learning models can be developed for the data obtained through swarm intelligence, and detection procedures can be created in connection with the behaviour of the elements oft he entities. They can be used to identify an abnormal traffic situation or an obstacle that affects traffic. A vehicle whose behaviour is special in some way (eg a towed vehicle, a slow or stationary vehicle, a slippery or poorly visible road section, a driver whose behaviour is strange or dangerous eg. due tosickness, driving on the road, maintenance vehicle in traffic, etc.). Among these, it is not always possible to recognize a vehicle on its own (in time), but this can be solved by cooperation between vehicles cooperating in an ad-hoc manner (occasional swarm). In the framework of the research (PhaseI.), the first possible use-cases can be identified, and the issues that can be solved by swarm intelligence. As a result of the project, a monitoring system will be set up, apart from the manufacturers of automated vehicles, in which the interoperability of individual vehicles can be examined independently. It will be possible to monitor the effect of automated vehicles, to what extent they change the reactions and driving habits of drivers sitting in human driven vehicles.

Innovative urban traffic management

In the framework of the RAJ project, a new type of road traffic light control system is planned to be installed (in phase II.). The aim of this prototype is to provide a flexible programmable traffic light controller device with a camera based data collection, and I2V communication technology, which can serve both road safety and real road AV tests. Such controller has not yet been established in Hungary.

The prototype signal controller – outside the ZalaZONE test track – could also be to temporarily controlled from ZalaZONE's central traffic control while testing. As it is an open place, the security functions are available like at other traditional traffic management equipment, the ZalaZONE control application will not have access to these safety-critical settings. The traffic light controller equipment should only be programmable via an API (Application Programming Interface). The functions provided are: classical monitoring functions; actual status; “on-line” managing period times, traffic light groups.

Conclusions

We expect a number of benefits with the appearance of automated vehicles in transport. Autonomous vehicles eg. can significantly increase road safety, reduce the number and severity of accidents involving personal injuries and make traffic smoother. In order to provide the appropriate infrastructure for AVs, additional efforts are needed. We have to ensure an IT data exchange platform to facilitate cooperative transport. Such an IT environment supports the traffic and testing of self-driving vehicles, regulatory activity, and the ability to change traffic rules, implementing a solid basis for conducting road tests.

‍

(The RAJproject is supported by the Innovative Mobility Program of the Institute ofTransport Sciences (KTI).)

‍