Architecture for high‐speed evolutionary computation

Genetic algorithms (GA) are widely known as a general-purpose optimization method, which can provide suboptimum solutions for various optimization problems by means of modeling genetic evolutionary processes of creatures. Several essential difficulties exist in GA, however, with regard to large amount of computation time and proper adjustment of many GA parameters. In order to overcome the difficulties of GA, this paper describes the architecture for a high-speed evolutionary computation, which is optimized hierarchic pipelines to take generation models into consideration and can be flexible genetic operations corresponding to a given problem. Simulation results evaluating the proposed architecture are shown to achieve a speed 193 times faster on average compared to software processing. © 2004 Wiley Periodicals, Inc. Electr Eng Jpn, 147(2): 39–52, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/eej.10308