A Buffer Length Selection Method for Systems with Random Measurement Delays

Due to the impact of environmental factors, external measurement in unmanned vehicle navigation often exhibits random time delays. Modified Kalman filtering with a buffer is commonly used to handle these delayed measurements. However, while increasing the buffer length can improve estimation accuracy, it can also significantly increase the computational burden and affect the real-time performance of the algorithm. Since the degree of observability can assess the ability of measurement information to observe the system state variables, this paper proposes calculating the degree of observability for different buffer lengths under different random delay conditions and selecting an appropriate buffer length based on the value of degree of observability.