Engineering Networks for Vehicle-to-Vehicle Communication

Kumar PortraitIt is often said that today’s science fiction is tomorrow’s science; however, this forward momentum in the advancement of technology does not happen by accident. As Thomas Edison famously stated, “genius is one part inspiration and ninety-nine parts perspiration.”

Texas A&M faculty member, Dr. P. R. Kumar’s, extensive research in control, communication, computation and manufacturing is a prime example of the level of commitment and hard work it takes to achieve something truly groundbreaking. Kumar, holder of the Engineering Chair in Computer Engineering and professor in Texas A&M’s electrical and computer engineering department, has been working on the emerging area of Cyber-Physical Systems, which are the “are the next generation of engineered systems in which computing, communication, and control technologies are
tightly integrated.”

In April of 2008, the BBC published an article highlighting vehicle collision-avoidance systems based largely on Kumar’s research in the field of wireless networks. Traditional networks that can only share information that has already been captured and stored, may soon be replaced by “embedded sensor networks,” which allow real-time information to be communicated instantly. In applications, this technology would allow, for example, vehicles to communicate with one another to monitor their speed and position.

Kumar and his group have developed a Cyberphysical Systems Lab at Texas A&M. Instead of working with full-size vehicles, Kumar and his students are using radio-controlled cars for beta testing the technology. The radio controlled cars are color-coded with patterned roofs to distinguish between them.

The cars are also rigged with cameras that supply real-time video of the environment surrounding the vehicle to a data server. The server is in turn able to communicate pre-programmed responses based on the information being received from the vehicles. This feedback loop between the vehicles and server is what allows them to navigate obstacles and avoid collisions as they occur in real-time.

The algorithm functions much the way our brains do when processing sensory input from our eyes. When the space in our immediate periphery is clear we know we are safe to take a step forward. Likewise, the car receives a proverbial “green light” when the area directly in front of the car is clear. On the other hand, if the cameras detect an obstacle, the car receives a signal to slow down or stop depending upon the proximity of the object. In the preliminary tests performed by Kumar and his team, up to eight cars interacted with each other in complex traffic situations, including parking, without colliding or otherwise interfering with normal traffic. The collision-avoidance systems, while promising, are not yet foolproof. Kumar acknowledged that the technology would need to accommodate objects outside the network, such as pedestrians and animals. The roadblock to largescale production lies in the complexity of design and operation, however, Kumar is optimistic these systems will one day simplify daily activities. Another area of research Kumar is studying is traffic intersections. In what they are dubbing “intelligent intersections,” Kumar and his students are replacing traffic lights and stop signs with interaction over a communication network of nodes including the intersection and the cars. Kumar notes that these types of intersections can help avoid waste created by unnecessary stops at rural intersections and potentially help deter accidents caused by human-error.

With the ubiquity of integrated computers within modern systems, the application potential for networks like these are virtually endless. From smart climate-control systems to fire-alarms to detection of moisture or even sound, these systems could be used in a number of commercial, industrial and environmental settings. Farmers can use automated moisture-sensing networks to water their crops to ensure they are not over or under-watering. Buildings may be able to sense temperatures and occupancy and render their operations more efficient. All of this is of course still theoretical, but, according to Kumar, “the Federal Communications Commission has already allocated a spectrum for vehicular networks.” “So things are happening fast, we are very much in this type of world already.”