Data/Information fusion; Optimal nonlinear estimation, Robust vehicle tracking and navigation; Situational awareness
Dr. Crassidis’s research involves the development of generalized multiple-model adaptive estimation (GMMAE) strategies that are used to provide significant convergence rate improvements of system model parameter and state identification, as well as enhanced probabilistic assignments of outcomes. Specific extreme event applications involve the use of GMMAE to provide risk assessment tools for situational refinement using both sensor and contextual data inputs. As Associate Director of the Center for Multisource Information Fusion (CMIF), Dr. Crassidis is responsible for level-1 fusion applications, which involve observation-to-track association, continuous state estimation (e.g. kinematics) and discrete state estimation (e.g. target type and ID) and prediction. Currently, a consortium between the University at Buffalo through CMIF, Texas A&M University and Virginia Tech is being conceived with the ultimate goal of mitigating disaster response coordination through the use of robotic sensors and uninhabited air vehicles in conjunction with state-of-the-art data fusion techniques.