University Events

Description: Topic: COMPARING THE ROBUSTNESS TO PARAMETER MISMATCH OF INFOTAXIS AND MAP SEARCH STRATEGIES

Location: Lester W. Cory Conference Room, Science & Engineering Building (SENG), Room 213A

Zoom Conference Link: https://umassd.zoom.us/j/99998476640
Meeting ID: 999 9847 6640 Passcode: 688944

Abstract:
Conventional target search algorithms can be categorized as either exploratory or exploitative. Exploratory methods prioritize comprehensive space exploration without incorporating acquired information. These approaches are well suited for high-clutter environments where individual detections hold low confidence. In contrast, exploitative algorithms heavily rely on existing data to refine their search. They excel in low-clutter scenarios where detections possess high fidelity. This thesis delves into a comparative robustness analysis of the performance between exploitative search strategies, specifically Maximum A Posteriori (MAP), and an advanced search strategy known as Infotaxis, which integrates both exploration and exploitation strategies. For this purpose, our focuses on assessing the robustness of MAP and Infotaxis under conditions of parametric mismatch during the Bayesian update and measurement processes. Parametric mismatch refers to the error conditions when the true probability of detection (PD) and probability of false alarm (PFA) utilized in the measurement function deviates from the model assumed estimated probability of detection (PD) and estimated probability of false alarm (PFA) in the Bayesian update function (BUF). Parametric mismatch conditions arise due to incorrectly calibrated computational models, impacting system accuracy and reliability. Numerical simulations consisting of 1000 Monte Carlo trials demonstrate that Infotaxis outperform MAP. These simulations conducted under varying search area size illustrate the superior efficacy of Infotaxis in comparison to MAP, showcasing its robustness and effectiveness in target search.

Advisor:
Dr. John R. Buck, Chancellor Professor, Department of Electrical & Computer Engineering, UMASS Dartmouth

Committee Members:
Dr. Ana Doblas, Assistant Professor, Department of Electrical & Computer Engineering, UMASS Dartmouth;
Dr. Paul J. Gendron, Associate Professor, Department of Electrical & Computer Engineering, UMASS Dartmouth

NOTE: All ECE Graduate Students are ENCOURAGED to attend.
All interested parties are invited to attend. Open to the public.

*For further information, please contact Dr. John R. Buck via email at jbuck@umassd.edu