Ultra - Wideband - based Trajectory Estimation Algorithm with Adaptive Square Root Cubature Kalman Filtering

Abstract

The accurate and real-time performance of trajectory estimation directly affects projectile observation and correction. However, estimation performance may suffer from poor real-time and weak anti-jamming capabilities when sensors, such as optics/acoustics, global positioning system, or radar are used to estimate trajectory. This study proposed a novel trajectory estimation algorithm based on the ultra-wideband (UWB) positioning technique to improve real-time estimation accuracy. The state equation was established based on UWB positioning technology, which combined time difference of arrival (TDOA) and angle of arrival, and then compared with single TDOA location technology. Square root cubature Kalman filter (SRCKF) was used to estimate the trajectory. However, the filtering accuracy decreased when the measured value was abnormal. The adaptive square root cubature Kalman filter (ASRCKF) method was proposed to deal with this problem. Judgment as to whether the ASRCKF method should be used was presented. Results show that the joint positioning method is superior to the single positioning method, and that the ASRCKF method can remove bad measured values. Hence, the method effectively solved the error that occurred in the measurement process and obtained accurate estimation results.