Los Angeles, CA
March 31, 2009 to April 2, 2009
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CSIE.2009.177
An algorithm based on grey innovation model GM (1, 1) of fixed length is introduced for the localization and tracking of moving targets. Kalman filter is an efficient computational method for tracking, but motion and noise assumption limits its process model to constant velocity model or constant acceleration model. The grey system theory uses the data characteristic of extrinsic randomicity and holistic regularity to find out the intrinsic rules of the system. It explores the law of subject’s motivation by accumulation of raw data and builds up the differential equations to estimate the next states of the system. Therefore an object tracking algorithm based on grey innovation model GM (1, 1) of fixed length is proposed and studied in detail. The effectiveness and efficiency of the proposed method is revealed through the performance comparison of grey innovation model and Kalman filter with constant acceleration model. A further study advice is discussed at the end.
grey innovation model GM (1, 1), localization and tracking, state estimation, Kalman filter
Xiao Yunshi, Yin Huilin, "Object Tracking Algorithm Based on Grey Innovation Model GM (1, 1) of Fixed Length", CSIE, 2009, 2009 WRI World Congress on Computer Science and Information Engineering, CSIE, 2009 WRI World Congress on Computer Science and Information Engineering, CSIE 2009, pp. 615-618, doi:10.1109/CSIE.2009.177