Sygic introduces driving behavior analysis to its mobile and navigation SDK

Sygic introduces driving behavior analysis to its mobile and navigation SDK
Published: 9 October 2019 - 6:15 a.m.
By: PMV Staff

Automotive navigation systems developer Sygic has introduced a driving behavior algorithm to its software development kit (SDK). The algorithm evaluates the user’s driving style in real-time with four variables: speed, acceleration, braking, and cornering.

Additionally, a distraction factor is taken into consideration, analyzing whether the driver is touching or using a mobile phone while driving. The more carelessly mobile app users drive, the lower will be the their score.

The algorithm uses phone sensors combined with unique technology developed by the Sygic team to get all inputs even without an internet connection. This allows for easy integration into any automotive or fleet solution. Cars with compatible connectivity, e.g. Smart Device Link (SDL), can even get access to built-in sensors in the car for an enhanced experience and reliability.

After each trip, the score is evaluated and saved in the Trip Log and can be sent to the backend and is used as a source for big data analytics, machine learning, and data mining.

The Sygic driver behaviour analysis will soon be available as an API for 3rd party developers of fleet management and insurance solutions. All modules extend the existing Sygic Mobile & Navigation SDK platform and Sygic Maps modules to make them available for third-party development teams.

Martin Strigac, CEO, Sygic, said: “The driving behaviour analysis can process data in real time and provide prompt feedback to the driver to improve his style while driving to reduce dangerous behavior proactively. Instead of late reports or dashboard with the trip log, we can educate and encourage them in a fun way by integrating our enterprise gamification features such as coaching, rewarding or social elements.”

Click here to add your comment

Please add your comment below
Your email address will not be published