Safer driving in a high-tech world

With today’s in-car gadgetry it is plausible that a driver could receive a phone call, a collision alert, a low-fuel warning, an instruction to turn right, and an internet stock update at the same time.

There is not, however, an in-car method for managing the information a driver receives but researchers at the US Department of Energy’s Oak Ridge National Laboratory hope to rectify this situation with the development of an In-vehicle Information System (IVIS).

Researchers at ORNL’s In-Vehicle Information System Development Centre have so far studied 36 men and women as they ‘drove’ in a simulator. The driving scenario involved bombarding the drivers with information from a cell phone, a collision-warning device, a navigation system, and an Internet-equipped computer screen.

Phil Spelt, principal investigator at ORNL’s In-Vehicle Information System (IVIS) Development Centre said there were a few crashes; one out of six of the drivers missed a turn during one of their 10-turn experimental runs; the majority could not remember much of the unmanaged general information they had received.

Spelt and his colleague, Daniel Tufano, are using this data to develop an IVIS that will manage competing bits of information being directed at drivers.

ORNL’s IVIS would filter phone calls, checking them against a database and passing them on to the driver only if they came from a known source. Information such as navigational cues would have precedence over phone calls, and the driver would receive prompts on where to go before being alerted to the phone call. Safety warnings, such as collision alerts, would have the highest priority.

Development of an effective IVIS system is said to be a two-fold challenge. From a technological viewpoint, the ORNL researchers’ challenge is to develop a network robust enough to manage multiple systems, but simple enough to be integrated with simple 8-bit systems such as pagers.

The IVIS team is currently studying ‘data-bus architecture’ which connects the disparate information sources through the network. There are currently two methods of integrating the information: the ‘bus-master with slave devices’ and the ‘peer-to-peer’ approach.

The bus-master forces all incoming information from slave devices through a filter, prioritises it, and displays only what was relevant in order of precedence.

The peer-to-peer system has no inherent hierarchy; devices communicate with each other and the driver with no prioritisation. Its flexibility, however, is said to permit a bus-master to be added at any point, should the drivers decide they need more control over information displayed in the car. Spelt sees the peer-to-peer information with optional bus-master as being the most likely answer to the problem for the immediate future.

The second, and perhaps most difficult challenge faced by the ORNL team is the issue of human interface. The IVIS designers must find a way to enable large amounts of information to be sorted and passed on to the driver in a way that will facilitate a safe and effective response.