Abstract:Two transition correction functions are proposed for the interaction multiple models (IMM) to adaptively update the Markov transition probability matrix (TPM) according to the situation. TPM is usually considered a constant as an initial value in a traditional IMM estimator. Fixed TPM is conveniently calculated by fixing prior information. However, inaccurate TPM can make the target state estimation error large. To solve the problem, the IMM algorithm must have time-varying transition probabilities such that the system model changes immediately according to the target movement. In this paper, two correction functions are designed by dividing them into two phases whether sub-filter model jumps occur or not. The first phase is the polarizing transition probability correction function, which increases the probability of the sub-model matching the target movement and decreases unmatched sub-models when the system does not transition into that particular sub-model. The second phase is the activating transition probability correction function, which quickly updates the probability when the sub-model jump occurs. We integrate the developed correction functions into the IMM algorithm in order to enhance its performance. Through the performance comparison between the proposed algorithm and others, the effect of probability correction functions is confirmed in simulations. Also, the flight experiment of fixed-wing UAVs demonstrates performance improvement.

Interacting multiple model algorithm based on the current statistical model

Multiple Model Algorithm Based on Adaptive Interaction of Model Set

Interacting Multiple Model Algorithm Based on Adaptive Transition Probability

Interacting Multiple Model Algorithm Based on Improved Current Statistical Models

Interacting multiple model algorithm based on H∞ filter

Hierarchical Interacting Multiple Model Algorithm Based on Improved Current Model

An Adaptive Multi-rate Interacting Multiple Model Algorithm

Half Adaptive Interactive Multiple Model Tracking Algorithm of Curvilinear Model

An Improved Imm Algorithm Based On Maneuvering-Adaptive Model Set

Improvement Interacting Multiple Models Algorithm

Hierarchical adaptive interacting multiple model algorithm

A Variable-Structure Imm Filtering Algorithm Based On Current Statistical Model

Adaptive interacting multiple model algorithm for high maneu-vering target tracking

An Improved Interacting Multiple Model Algorithm With Adaptive Transition Probability Matrix Based on the Situation

Interacting Multiple Model algorithm with Quasi-Monte Carlo Kalman Filter

Interacting Multiple Model Algorithm for Tracking of Maneuvering Targets

Interacting multiple models algorithm based on α-β & α-β-γ filters

Target tracking using the interactive multiple model method

Double Markov Multiple Hypothesis IMM Maneuvering Target Tracking Algorithm

IMM Algorithm Based on H∞ Filter for Maneuvering Target Tracking

A Novel Interacting Multiple Model Algorithm Based on Multi-Sensor Optimal Information Fusion Rule