Discrimination of moving status for logistic tracking systems using Mahalanobis distance of acceleration record

An algorithm that judges the moving or staying statuses of distribution equipment by continuous measurement of the vibration is developed to realize a low-cost and power-saving logistic positioning systems. First, the measured acceleration levels are divided into two groups using tentative threshold level. Then, the threshold level is updated repeatedly using Mahalanobis distances of two groups to minimize the probability of discrimination error. Also an approximate formulas for the threshold to reduce computation amount is presented. Finally, to confirm the effectiveness of this method, the discrimination accuracy was evaluated using the simulation and actually measured data of an automobile.