Minimax particle filtering for tracking a highly maneuvering target

Abstract

Summary In this paper, we propose a new framework of particle filtering that adopts the minimax strategy. In the approach, we minimize a maximized risk, and the process of the risk maximization is reflected when computing the weights of particles. This scheme results in the significantly reduced variance of the weights of particles that enables the robustness against the degeneracy problem, and we can obtain improved quality of particles. The proposed approach is robust against environmentally adverse scenarios, particularly when the state of a target is highly maneuvering. Furthermore, we can reduce the computational complexity by avoiding the computation of a complex joint probability density function. We investigate the new method by comparing its performance to that of standard particle filtering and verify its effectiveness through experiments. The employed strategy can be adopted for any other variants of particle filtering to enhance tracking performance.

Noise-Adaption Extended Kalman Filter Based on Deep Deterministic Policy Gradient for Maneuvering Targets

“…With the development of high maneuvering targets, such as hypersonic aircrafts and missiles, maneuvering target tracking has recently drawn a lot of attention [ 1 , 2 ]. During the past few decades, lots of tracking methods have been developed and investigated for maneuvering targets [ 3 , 4 , 5 ]. Among the existing target tracking approaches, designing an appropriate model for the target maneuvering motion is regarded as an important part of decreasing the tracking errors caused by the model mismatch.…”

Tracking by Risky Particle Filtering over Sensor Networks

“…In this paper, we propose minimax-PF (MPF) for the target tracking based on multi-RSS-measurements at up to ten sensors. The minimax PF approach was initially introduced in Reference [34] recently, where a maneuvering target is tracked based on the range and bearing, which are only two measurements; therefore, the methodology was not fully investigated in terms of the number of measurements, clarity of the derivation of the approach, analysis of risk and weight concerning to all compared variants of PFs. The investigation of the method in the application of networked sensor measurements where a large number of sensors are available is more appropriate for the verification of the approach.…”