Smartphone Based Real Time Riding Behavior Detection

With the drastic increase in the number of shared bicycles in the majority of the Chinese cities, there is a need for implementation of such a system that can monitor individual rid-ing behavior of the rider. This way an awarding and punishing system can be implemented by adding or removing bonus points from the rider's profile. This incentive encourages the bicycle riders to use the bikes in a careful and safe way, thereby increasing the rider's safety and decreasing the loss of the company in terms of recklessly used bicycles.This dissertation gives the elaborative information about Behave: A smartphone based real-time riding behavior detection application. Behave can be implemented as a module in the bicycle sharing apps to provide such functionality. It can accurately identify the speed, left/right turns, U-turns, brakes, dangerous brakes, left/right leaning and dangerous leaning, and dangerous bumps. In the case of dangerous riding behavior,it also issues a dangerous behavior warning to alert the rider. The research for improving the safety of bicycles in the field of Intelligent Transportation Systems (ITS) is quite scanty.Mostly, the researchers target to improve the safety of cars. Moreover, the researches that have been performed on bicycles have utilized individual sensors that are quite ex-pensive and difficult to equip for an average user. On the contrary, Behave just utilizes the smartphone sensors without the need of attaching any peripheral devices.