A New Adaptive Curve Model Tracking Algorithm and Its Application
Feng Zhang · 2019
A new adaptive filtering method based on curve model is proposed to solve the problem of large positioning error caused by the mismatch between maneuvering target modeling and actual target motion. The method in the paper abandons the traditional method which estimates the turn rate according to the direction angle and calculates the tangential acceleration based on the interpolation acceleration. The paper takes the turn rate and the target acceleration both as the state variables, expanding the state vector, which improves the estimation accuracy of tangential acceleration, and alleviates the large computational burden caused by the traditional two-layer filter structure. The expression of process noise covariance after state vector expansion is derived. In addition, the design of direction angle is optimized based on the arc tangent library function "atan2()" in the embedded software, combined with the transfer relationship of direction angle between four quadrants. Through simulation, it is proved that the proposed method is more adaptable to maneuvering target tracking and has higher positioning accuracy.