标题：Heuristic Kalman Algorithm for Multiobjective Optimization. * * This work has been partially supported by the National Council of Scientific and Technological Development of Brazil (CNPq) through the grants 303908/2015-7-PQ, 304066/2016-8-PQ and BJT-304804/2014-2
摘要：AbstractThis papers presents a new heuristic multiobjective optimization procedure based on Kalman filtering. The novel approach is compared to the well-known Nondominated Sorting Genetic Algorithm version II with the Zitzler, Deb and Thiele test suite showing favorable results for the Kalman based solution. Further improvement on the proposed algorithm is made with respect to the learning factor, which is set to be time-dependent, improving the algorithm in most cases and diminishing the number of control parameters. The proposed multiobjective optimization algorithm is interesting due to its convergence and coverage quality, its computational complexity and few parameters to be adjusted.
