The BME Automated Drive Lab delivered one of the professional sensations to the ITS World Congress, held in Hamburg, in 2021. October 11-15. Carried out on different continents each year, it is the most significant global event within the intelligent transportation industry. At the same time, participating in the event equals the entry to the international professional community.
This year, more than 200 industry representatives, like research institutes, traffic infrastructure operators and organizations from the private sector participated in the conference from all parts of the world. Despite the pandemic, they showcased their solutions and projects to more than 10 thousand people.
Hungary has been the participant of this international event for the first time, which proves the dynamic development of the Hungarian intelligent transportation industry and that we have R&D research results that belong to the global forefront. This is what the presentation of the BME Automated Drive Lab confirmed, hand in hand with two of its strategic partners, the Hungarian Public Road PLC and ZalaZONE.
Our imposing stand was increasingly popular, where we carried out a special demonstration. Besides the presence of Professor László Palkovics, the Minister of Innovation and Technology, the renowned international representatives of the industry commented approvingly on our virtual demonstration and the research results.
Our Lab has been pursuing a tight cooperation with these partners for a long time now. Our common goal is to make the Hungarian public road system intelligent, to fortify the Hungarian research & development and to actively connect Hungary into the international professional ecosystem.
“The introduction proved to be impactful.” - summed up the event Dr. Zsolt Szalay, the Founder and Director of the Automated Drive Lab. He has been leading the team that introduced the Cloud Based Digital Twin and Smart Infrastructure System in Hamburg. The live demo of the system validated the whole concept. “We received positive feedback only. We found that the approach represented by the system has a global relevance, while it also turned out that we are the first ones on a global scale with this revolutionary solution.”
The central element of the concept is the digital twin - already widespread in the manufacturing industry -, which materializes by processing in synchrony the data of the environment and the vehicle. Self-driving is mainly associated with those vehicles that are endowed with sensors. However, if we set the environment up with adequate sensors and we connect the data collected by these devices with the data collected by the vehicle, we get an ecosystem that can be construed in a complex way. Within this ecosystem, it is not only the self-driving car that gets a notification or a message, but every moving element, like a pedestrian crossing the road.
The synthesis of the information comes to fruition in a digital twin created within a cloud infrastructure. The moving parts receive the information from this digital twin that is being refreshed in real-time. Thus, the intelligent (self-driving) vehicles’ visual range is increased from the current few hundred meters to even 5-10 kilometers, as it is able to calculate events occurring kilometers away from it, thanks to the digital twin.
In order to achieve this, the digital twin needs to receive and model every change real-time. First of all, we need to create a static HD map with centimeter accuracy based on a pointcloud (including all important permanent elements, like the roadside, benches by the road, road paint, trees, bins etc.). Then, we need to administer the seasonally changing, semi-static parts (such as temporary road works), the semi dynamic elements, which are only relevant for a couple of hours, (accidents, a tree blocking the road) and the dynamic pieces of information (pedestrians crossing the crossroads). All of it in real-time. Thanks to the information deriving from this technology and the digital twin, the car’s system is able to recognize in due time if someone just stepped on the crossroads the car is about to approach.
The audience had the pleasure of experiencing and trying out this globally new solution at the ITS World Congress, within the boundaries of a real-time demonstration. According to Dr. Zsolt Szalay, the team could successfully convey the message of the Cloud Based Digital Twin and Smart Infrastructure System; the sensor systems and the cloud-based digital twin of the Automated Drive Lab can visualize and interpret a given road segment real-time with all its changes.
In order for the project to be feasible, the team had to utilize numerous modern technologies. The road segment (crossroads) is monitored by sensor-stations containing various sensors (optical cameras, infra cameras, LiDARs, radars, temperature sensors etc.) that detect and note the changes. (For manageability reasons, two sensor-stations have been constructed in Hamburg for the demonstration). An AI based system is responsible for the interpretation of changes, like detecting the pedestrians and tracking their movements. To achieve flawless operation, AI algorithms synchronize the signals coming from the sensors. Last, but not least, the system needs low-latency data connection, which can be provided by 5G infrastructure. Due to these technologies, if a car is approaching a crossroads, the car’s systems receive information real-time about the movement of the pedestrians, right from the cloud-based digital twin. As a result, it can stop when someone steps on the road.
This is the technology that the audience of the ITS World Congress could see in Hamburg in full-swing. If someone entered the booth and walked across the demo crossroads, the approaching (simulated) vehicles stopped in the virtually created digital twin. The vehicles only continued their way forward, when the person left the crossroads. The demo has been made even more life-like with the use of an Augmented Reality environment: thanks to AR glasses, the members of the audience could feel as if they were in a real-life situation and saw, as the car stopped in front of them when they stepped on the crossroads.
After the presentation at the ITS World Congress, several players from the automotive industry have shown interest in the Cloud Based Digital Twin and Smart Infrastructure System, while it soon will be put into practice in Hungary. Sensor-stations will be placed to seven crossings of the ZalaZONE Smart City and the technology will be used on a 1,5 km long test motorway. Meanwhile, on Hungarian motorway M76, based on the specifications of the Lab and ZalaZONE, they will construct a 12-km-long intelligent road, which will encompass 75 sensor-stations, complemented by sensors built into the road (cameras, LIDARS, radars, temperature sensors, crossing sensors, weight sensors etc.). Moreover, the plans include intelligent parking lots, variable road signs and communication interfaces, among others. The road segment will be used for testing purposes. However, the research doesn’t stop here, as exploring the swarm-like behavior of the traffic proves to be interesting for the team. For this purpose, five sensor-stations will be placed soon on the Hungarian M1-M7 motorways’ entry passage.
In the following video we introduce the system developed by the BME Automated Drive Lab team, while you can read the related professional studies here.