Research On Online Vehicle Trajectory Compression Method Based On Motion Behavior Pattern Recognition Of Smart Phone Sensors

With the popularization of various portable mobile devices with positioning function,a large number of spatio-temporal trajectory data of moving objects have emerged,and the huge data scale has brought severe challenges to trajectory data management and analysis.Facing a large amount of trajectory data,there is an urgent need for an efficient trajectory data compression technology.The current compression algorithm for trajectory data is usually the original trajectory that has been sampled by high frequency in advance.If these algorithms are applied to mobile terminal devices that collect trajectory data(such as smart phones,etc.),it will undoubtedly cause a huge burden on the storage space and energy consumption.In addition,there is still room for improvement in the real-time performance of existing compression algorithms,so that they can be better adapted to network transmission of data.Aiming at the above shortcomings,this paper proposes an online compression algorithm for spatio-temporal trajectory data based on vehicle motion behavior recognition,which achieves the goal that the collected trajectory points are feature points.The algorithm recognizes the steering behavior and speed change behavior of the vehicle by monitoring and analyzing the data change law of the linear acceleration sensor and direction sensor built in the smartphone under different motion behavior modes of the vehicle,and uses satellite positioning sensors to record the corresponding trajectory feature points according to the motion behavior pattern recognition result,so as to realize real-time online data compression of vehicle trajectory.For vehicle speed behavior pattern recognition,this paper uses three time series stationarity test methods(ACF method,M-K method and Lunci method)to identify the uneven data segment of the collected linear acceleration sensor data.Through actual verification,it is found that the M-K test method has the best high recognition rate.For vehicle steering behavior pattern recognition,this paper uses several steps such as steering threshold to identify the starting point,cumulative steering angle acquisition intermediate point,time threshold to control sampling density,and steering amplitude to refine the steering behavior mode to effectively complete the recognition of vehicle steering behavior patterns.In order to comprehensively verify the compression performance of the algorithm in this paper,the proposed algorithm is compared with the representative feature points extraction based trajectory compression algorithms under four road conditions.And the results indicate that although it is slightly weaker than the representative trajectory compression algorithms in compression accuracy,it can compress the trajectory data online,with strong real-time performance and high calculation efficiency,which reduces the amount of network transmission data.It only requests positioning and sampling at key feature points,which leads to the algorithm being able to adapt to changes in road conditions to a certain extent,reducing the storage space of the mobile phone,and decreasing the power consumption of the mobile phone to a certain extent.