- AI radar monitoring starts test measurements in Mönchengladbach
Around the main station in Mönchengladbach, test measurements are being conducted using radar technology that could potentially assist police surveillance in the future with the help of Artificial Intelligence (AI). A total of eight radar sensors, four in the station's entrance hall and four at the Platz der Republik, are recording movements and actions using color spectra and wave lines.
AI is supposed to recognize security-threatening situations from the recordings and alert the police to a potential threat in the future. "During the test phase, the system will be trained for three scenarios: escape, physical altercation, and a helpless person on the ground," explained Thomas Patalas, a scientific employee at the University of Applied Sciences Niederrhein, which is scientifically accompanying the project called "KIRaPol.5G".
Police supports project
For this, AI must first learn what the recorded patterns mean. Therefore, actions are planned around the station where people will reenact certain scenes. "Then we can see if the AI reacts and if the sensors are capturing at the right angle," said Patalas. However, the technology is not yet connected to the police, but the police is supporting the project by providing training and video material that AI can learn from.
In order to evaluate the recorded radar data and assign them to an existing or non-existing danger, video data is also recorded in parallel. In addition to scenes reenacted at the police training center in Linnich, the project team can also access the video cameras of the federal police in the station's entrance hall, explained Patalas.
The goal of this innovative technology is to increase security while also protecting the privacy of citizens. "The advantage of AI-assisted radar technology is that the intrusion into the personality and data protection rights of the people present in a observed scene is estimated to be very low compared to video surveillance," said Patalas.
Radar technology has advantages
The sensors only record anonymized data from which no personal identity can be inferred. Moreover, the technology has another advantage: radar is insensitive to light conditions and environmental influences such as fog, rain, or smoke. The applied radar technology is already in use at other locations and is used, for example, for the development of autonomous driving.
The current measurements are planned to continue until about September, according to Patalas. Afterwards, the data from the project, which is funded by the state of NRW until the end of the year and involves a total of five partners from industry, science, and security authorities, will be evaluated. Only then can a decision be made on how and whether the system could also be used in real-time operation at train stations and public places in the future to provide more security.
The AI's ability to recognize security threats relies heavily on its understanding of the recorded patterns, which it can only gain through learning and training. Therefore, the project involves staging scenarios around the station for the AI to analyze. Additionally, the enhanced privacy aspect of radar technology compared to video surveillance is one of its key selling points, as it only records anonymized data.
