SIEMENS

The driver of the tanker truck doesn’t know that he’s heading for disaster. He’s unaware that the braking system on one of his rear wheels is blocking and beginning to glow red hot. There’s a tunnel coming - in three kilometers - but the potential catastrophe doesn’t have a chance to unfold thanks to newly developed safety systems that have already detected the rolling time bomb and triggered an alarm in the tunnel operator’s control center. Here, staff switch the lights at the tunnel entrance to red, and flashing hazard signs redirect the driver in order to defuse the dangerous situation.

This scenario is still a future vision. Nevertheless, a research project known as SKRIBT - (German acronym for “Protection of Critical Bridges and Tunnels on Roads”) - which is being conducted by scientists at Siemens Corporate Technology (CT) and its Mobility Division, is moving closer to making this vision a reality. Ten partners from government agencies, industry, and research institutes are participating in a threeyear project, which is being funded by the German Ministry of Education and Research. The aim is to make critical road segments safer. “Tunnels and bridges are the most important components of the road network,” says Dr. Frank Heimbecher, project coordinator at Germany’s Federal Highway Research Institute, which initiated the SKRIBT project. “If they get damaged, the consequences can be economically devastating.”

Most major accidents in tunnels involve defective trucks - situations in which tires blow, Molecular Detectives | Tunnel Safety brakes overheat, or engines fail in a manner that triggers a fire. That’s why Alla Heidenreich, infrastructure project manager at Siemens CT, has been working with her team since 2008 on two safety systems that can identify defective trucks and those transporting hazardous materials - before they enter a tunnel. The researchers, who are from Munich and Princeton, New Jersey (USA), came up with the idea of combining video images with thermal imaging technology. This enables them to determine if certain vehicle components are overheating. The system works as follows: A video processing program linked to surveillance cameras identifies a passing truck and converts a segmented two-dimensional image of it into a 3D model with using newly-developed algorithms. The program is then able to recognize components susceptible to fire, such as wheels, brakes and axles.

The thermal image of the truck, which is recorded using an infrared camera, is linked with the 3D image, after which an analysis program searches for anomalies that could indicate defects. It does this using knowledge gained from models that provide information on things such as how hot one axle may get in relation to the others. Because normal video cameras need expensive external lighting at night, Siemens researchers are working on yet another idea. “Our next step will be to study possibilities for the exclusive use of infrared images to identify potentially dangerous situations with tires, brakes, and axles,” says Dr. Andreas Hutter, an expert in realtime image processing. “If we succeed, we’ll be able to significantly reduce costs.”

Hazardous material transports pose an even greater problem, especially if it’s not clear what type of cargo is being shipped. Some materials like gasoline may only be transported by truck through certain tunnels. Up until now, there has been no automated system for monitoring compliance with such rules. Trucks today are in fact required to carry orange stickers bearing coded information on how dangerous their freight is and which categories of tunnels they may pass through. However, video cameras can not decipher these labels when visibility is poor or the labels are covered with dirt. Radio Frequency Identification (RFID) transponders would thus offer a major benefit here.

Transmission-Enabled Stickers. If experts at CT have their way, trucks will soon also be equipped with hazardous material labels containing a small RFID chip that can be read via radio and that also holds all information about what the truck is carrying. “That would significantly increase the accuracy of the monitoring system,” says Heidenreich. Such a system would function roughly as follows. When a truck passes a reading point approximately three kilometers before a tunnel, its cargo data would be registered by the RFID system and forwarded to a control center. Only one truck transporting hazardous materials would be permitted in the tunnel at a time. Should an accident occur, firefighters would tackle the blaze using precisely the right extinguishing agent. Any truck attempting to enter a tunnel with prohibited freight would be stopped by a red light in front of the entrance.

The CT team is particularly proud of its newly developed RFID transponder system’s ability to meet extremely high demands. The chip can transmit its signal to the unit’s reading device over a distance of around 50 meters - and send the data at least twice within two seconds. “Conventional passive radio chips without a built-in energy source have a range of only six meters,” says Daniel Evers, an RFID expert at CT. “That’s why we use an active chip that has a built-in battery and transmits in the high-frequency range of 2.45 gigahertz. To ensure the battery lasts as long as possible, the transmitter in the transponder sleeps until it’s woken by a radio pulse issued by the reading device at the checkpoint.” Evers also points out that the RFID data cannot be intercepted or falsified. To ensure this is the case, Siemens researchers employ an encryption technique they previously developed for passive RFID chips (see article “Products that Don't Lie“, Pictures of the Future 1/2009). “Previous solutions needed too much energy,” says Hermann Seuschek, an IT security expert at CT. “However, our cryptochip is so energy efficient that the transponder can run for at least three years without needing a replacement battery.”

Research activities will be followed by road tests in mid-2010, when Siemens researchers will install truck detection system components at the Aubinger Tunnel near Munich. Plans call for the tunnel safety system to be tested until February 2011. “Up until now, activities have focused on improving safety within the tunnel,” says Heidenreich. “But in the future, we’re going to be able to detect and prevent danger before a vehicle gets there. Video, RFID, and infrared technologies will play a key role in this process.”