Journal of Global Optimization
Minimal representation multisensor fusion using differential evolution
CIRA '97 Proceedings of the 1997 IEEE International Symposium on Computational Intelligence in Robotics and Automation
GECCO '05 Proceedings of the 7th annual conference on Genetic and evolutionary computation
Hi-index | 0.00 |
Differential Evolution (DE) is an evolution algorithm that was proposed by Storn and Price in 1997, which has already succeeded in applying to solve optimization questions in a lot of fields. This paper discusses various kinds of characteristics that DE demonstrates at first, then propose an improved differential evolution algorithm (TSDE) which has three kinds of efficiently evolutionary strategies, and proves its global convergence property use the finite Markov chain theory. Through the compare result of calculation of 5 classics testing functions, it shows that TSDE has the obvious advantage in the quality of solution, adaptability, robustness etc. than original DE and DEfirDE.