An Optimized Algorithm For Trajectories Compression In Agricultural IOT

Global Positioning System(GPS)can provide position information for unmanned robots(UR),making automatic trajectory tracking possible.Due to the limitation in the sampling rate and the resolution of the GPS,outdoor agricultural UR is facing great challenge due to the speed requirement and the narrow terrain condition in Japanese and Chinese agricultural fields.Sensor fusion is one of the solutions to the existing challenge.By adding the sensor such as a gyroscope and an acceleration sensor with a fast sampling cycle in addition to GPS,positioning information(e.g.angular velocity)can be obtained to synthesize and estimate path-motion data.The objective of this study is to establish a measurement system based on INS/GPS sensor fusion to perform trajectory tracking at high speed and precision,as well as to provide adjustments to pre-determined trajectories.The integrated system should have positioning characteristics for continuous and high-precision positioning results.In case of GPS signal blocking,the INS kick in for a short period of time.Long-period GPS-led,short-time INS-assisted mode can reduce the instability significantly due to GPS off-line at short-time-scale,and corrects the cumulative error caused by INS at long-time-scale.In the meantime,the displacement and navigation angle data leads to accurate positioning of the robot and timely motion adjustments based on analysis of the system variables.In order to provide more accurate position estimation,especially more accurate direction angle estimation,local gravity field information is used to correct the system position error when the GPS is interrupted in the combined system.High-precision gravity deviation compensation not only improves the positioning accuracy,but also improves the accuracy of the attitude angle of the motion platform.