Research On Multi Target Tracking Algorithm Based On Doppler Radar

Doppler radar detects the position and relative velocity of moving targets by using the Doppler effect.Because of the particularity of its working principle,it is widely used in airborne early warning,navigation guidance,war detection and other fields.In recent years,with the continuous development of multi-target tracking algorithms based on stochastic finite set,how to give full play to the performance of Doppler radar under the stochastic finite set framework has gradually become a research hotspot.In this paper,multi-target tracking based on Doppler radar measurement information is taken as the research background,starting from the adaptive initiation of target track,the application of Doppler radar measurement,maneuvering multi-target tracking and Doppler blind area masking,etc.δ-Generalized Labeled Multi-Bernoulli(δ-GLMB)filter has been improved accordingly to adapt to various complex application scenarios.The specific research content is as follows:(1)A δ-GLMB algorithm for adaptive trajectory initiation is proposed to address the issue of insufficient utilization of measurement information in the δ-GLMB filter and the ability to only detect new targets at fixed positions.The algorithm introduces the Doppler radial velocity information in the Doppler radar,carries out measurement conversion and sequential filtering on the position measurement information and radial velocity information in order to improve the tracking accuracy.At the same time,it uses the velocity information implied in the radial velocity to calculate the state information and newborn distribution of the target,filters the measurement according to the speed rules and Doppler information rules of the observation target,and groups the measurement set by calculating the measurement correlation probability,To reduce the problem of short false tracks caused by clutter.Finally,simulation experiments have shown that the proposed algorithm effectively utilizes Doppler information and can achieve adaptive track initiation in the detection area based solely on measurement information,with better tracking accuracy.(2)To address the issue of basic δ-GLMB filters being unable to track maneuvering targets with varying motion states,a δ-GLMB algorithm based on interactive multiple models is proposed.This algorithm draws inspiration from the interactive multi model algorithm framework in single target tracking,using multiple motion models to describe the target’s operating state.Each model is transferred through a Markov probability transfer matrix,and soft switching between models is achieved through interactive output,solving the problem of filtering divergence caused by a single motion model not being able to fit the maneuvering target’s motion state.At the same time,in order to prevent the expansion of Markov jump branch size,a branch merging strategy was adopted to control the number of branches.Finally,simulation experiments have shown that the proposed algorithm outperforms the multi modelδ-GLMB algorithm based on branch pruning strategy in terms of tracking accuracy and runtime.(3)In order to solve the problem that the tracking performance of Doppler radar is deteriorated due to the Doppler blind area,a δ-GLMB algorithm incorporating the minimum detectable speed is proposed.This algorithm uses the minimum detectable speed to model the detection probability of radar,and incorporates this model into the δ-GLMB algorithm.During the update process,pseudo measurement and enhanced measurement components are added to ensure that the target track can still maintain a large weight within the blind zone,avoid being deleted by pruning thresholds,and redistribute track information after the target exits the blind zone to make the track continuous,thereby suppressing the impact of Doppler blind zone on the tracker.Finally,simulation experiments have shown that the proposed algorithm can effectively suppress the impact of Doppler blind spots and has stronger robustness when facing Doppler blind spots of different widths.