Stochastic Systems Group
Home Research Group Members Programs  
Demos Calendar Publications Mission Statement Alumni

SSG Seminar Abstract


Gaussian Mixture Reduction for Target Tracking


Jason Williams
SSG, MIT


The problem of tracking targets in clutter naturally leads to a Gaussian mixture representation of the probability density function of the target state vector. Multiple Hypothesis Tracking (MHT) techniques maintain the mean, covariance and probability weight corresponding to each hypothesis, yet they rely on ad hoc merging and pruning rules to control the growth of hypotheses. This research proposes a structured, cost-function-based approach to the hypothesis control problem, utilizing the Integral Square Difference (ISD) cost measure. The performance of the ISD-based algorithm for tracking a single target in heavy clutter is compared to previous methods, revealing a remarkable improvement in the average track life achievable.



Problems with this site should be emailed to jonesb@mit.edu