14 DAY TRIAL // Microsoft Research has developed a new technology that uses a smartphone connected to an OBD-II scanner to detect and warn of factors that could increase the risk of accidents, such as speeding, drunk driving, or distractions.
Specifically, the so-called HAMS (Harnessing AutoMobiles for Safety) consists of a smartphone that’s attached to an OBD-II scanner and connected to Azure to collect, analyze, and visualize data which can be then used to assess the risks of accidents for each individual.
The smartphone needs to be placed on the car’s dashboard, with the front camera monitoring the driver and the back unit looking at the road. The phone’s GPS is also being used to track the route, and so are inertial sensors which can determine unusual behavior from the car and the driver.
“Based on the raw data obtained from these sensors, we have built detectors for various events of interest including driver distraction, fatigue, etc. and also for vehicle ranging, to determine whether a safe separation distance is being maintained,” Microsoft explains.
10,000 km of data and counting
Redmond has already tested the system on a fleet of cabs at its own research center in Bengaluru, analyzing more than 10.000 km of data in just a few months, but further refining is still needed, as the firm itself explains.
Microsoft claims that such a system can be safely used on a fleet, such as cabs, buses, or trucks, where drivers can be easily monitored by their supervisors and get alerts whenever higher risks of accidents are detected.
“It is our thesis that improving road conditions, vehicle health and, most importantly, driver discipline would help boost road safety. Indeed, among the leading causes of road accidents are such factors as speeding, drunk driving, and driver distractions, all of which can be mitigated through better driver discipline. The key to bringing about an improvement is effective monitoring leading to actionable feedback,” the firm notes.
The system is still in development at the moment, but depending on how testing goes in the coming months, it could make its way to select fleets, the company hopes.
