Filter initialization method under condition of unequal measurement time intervals

Abstract

The invention discloses a filter initialization method under the condition of unequal measurement time intervals, and belongs to the field of radar data processing. A target is tracked by use of a Kalman filtering algorithm. A filter is initialized first, including filtering state initialization and filtering covariance matrix initialization. A filter initialization method based on three-time measurement difference has an obvious defect, namely, a filter cannot be initialized directly under the condition of unequal time intervals of measurement data, and the method can hardly adapt to interruption or loss of measurement data caused by problems like low signal-to-noise ratio and track initiation uncertainty. The filter initialization method under the condition of unequal measurement time intervals put forward by the invention intends to solve the problem, and is of vital significance to effectively enhancing the adaptability of the Kalman tracking filtering algorithm. The experimental simulation result of the embodiment shows that the method of the invention can effectively overcome the application limitations of the filter initialization method based on three-time measurement difference. Therefore, the filter initialization method under the condition of unequal measurement time intervals disclosed by the invention has high engineering application value and a broad promotion prospect.

Description

technical field [0001] The present invention relates to a radar data processing method, in particular to a method based on Kalman tracking filter (including Kalman filter, extended Kalman filter, measurement conversion Kalman filter and non-equal radar measurement data time interval). Min Kalman filter, etc.) filter state and filter covariance matrix initialization method, suitable for radar target tracking filter field. Background technique [0002] To use the Kalman filtering algorithm to track the target, the filter needs to be initialized first, including the initialization of the filter state and the initialization of the filter covariance array. When the radar is used to detect and track the target, the measurement data may be interrupted and lost due to the low signal-to-noise ratio of the target and the uncertainty of the track start, resulting in the time interval of the measurement data used for filter initialization. As a result, the filter cannot directly use th...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to propose a filter initialization method under the condition of unequal measurement time interval, and solve the direct initialization problem of the Kalman filter under the condition of unequal measurement data time interval

Images