[Legacy Report] CSS and SMCS Chapters Joint Seminar on An adaptive Online Co-search Method with Distributed Samples for Dynamic-target Tracking
CSS and SMCS Chapters Joint Seminar on An adaptive Online Co-search Method with Distributed Samples for Dynamic-target Tracking
Speaker: Feng Li, Ph.D. Candidate, Donghua University, Shanghai, China
Place: ECE 202
Time: 10:30am, Wed., September 21, 2016
Abstract: By the motivation to perform a search and rescue task, we clarify a dynamic optimal problem in a complex environment if the interested target unpredictably travels in an environment with general non-Gaussian distributed and time-varying noises, and recursive Bayesian estimation with distributed sampling (RBEDS) method followed by two kinds of communication coordination, which are real-time communication and communication after finding the target, constructed to solve this issue. To search the target effectively and efficiently, an online updating algorithm and a self-adaptive control protocol are designed to hit the balance between the exploitation and exploration. A series of simulation and experiments demonstrate that the local samples help searchers to keep flexible and adaptive to the changes of the target. Searchers with RBEDS can perform a comparable search performance with the global sampling method, Markova Chain Monto-Carlo estimation for instance, by applying a real-time communication. However, the proposed method with both two communication coordination exhibits a good accuracy when tracking the target. Another attractive result is that only a few searchers and local samples are demanded. The insensibility to the scale of the sample makes our method obtain a better solution with less computation cost.
Bio-Sketch: Feng Li received the B.Eng. degree in automation from Donghua University, Shanghai, China, in 2012, where he is currently pursuing the Ph.D. degree in control science and engineering with the College of Information Sciences and Technology. He is now a visiting doctoral student with New Jersey Institute of Technology, USA. His research interests include machine learning, intelligent control, and multi-robot systems.
Add Event to Calendar
