A new research project aims to integrate a motor vehicle's multiple microprocessors to create a "cognitive car" that can predict vehicle failures before they happen, help re-direct drivers to less congested routes, and help reduce traffic accidents.
These same capabilities could also be applied to commercial trucks as well, said Alan Wassyng, Ph.D., acting director of the software quality research laboratory at Canada’s McMaster University. The lab is heading up the project as part of a shared university research award from IBM.
“I see no reason why trucks should be different from cars in this regard,” he told FleetOwner. “There may even be more functionality that would be of benefit in truck.”
Engineers at McMaster will study how to link a vehicle’s entire electronic system through a single multi-core or “super” microprocessor that’s designed to perform many complex calculations simultaneously. IBM originally developed this “super” microprocessor for the video game industry, but is now using them in much wider applications within the financial services, energy exploration, medicine, and digital animation industries.
“To date, our research has focused on safety-critical software in industries such as nuclear energy and medical devices, but increasingly the automotive industry is adding functionality to vehicles that is safety-critical,” said Wassyng. “Investigating how a powerful multi-core processor could be applied to manage that functionality will go a long way in helping build smarter vehicles that help drivers operate more safely and efficiently.”
McMaster’s research team will focus on integrating data from sensors and microprocessors installed in both vehicles and roadways to give drivers more “real-time” visual information and alerts to avoid road congestion, helping improve safety while reducing emissions related to stop-and-go traffic.
The program will also study how the increased computing power can better integrate vehicles into regional and global transportation systems, including roadside service, traffic management, air quality management, and emergency services.
The key is creating what Wassyng calls an “overview” role for the super microprocessor.
“This would help us provide better information about the vehicle itself to the driver, and also information about the vehicle's immediate environment,” he said. “We would choose to provide alarms/warnings to the driver when possible, but may have to have some control possibilities also. For example, entering a highway and having another vehicle in your blind spot -- a warning may not be adequate. We may have to slow the vehicle down, prevent the steering action and sound an alarm.”
Wassyng added that detection of pedestrians and possible collision avoidance are other possibilities for this microprocessor control system , but stressed that figuring out how all of this impacts the driver will be critical to making it work.
“Our own emphasis in these cases is in ensuring safe behavior,” he said. “We humans tend to rush in with new technology. We think this technology has the potential to make vehicles much safer. Yet if this active safety software does not work correctly, it will be worse than not having it at all.”