论文信息 - A Dynamic Fitness Function Applied to Improve the Generalisation when Evolving a Signal Processing Hardware Architecture

A Dynamic Fitness Function Applied to Improve the Generalisation when Evolving a Signal Processing Hardware Architecture

Evolvable Hardware (EHW) has been proposed as a new method for designing electronic circuits. In this paper it is applied for evolving a prosthetic hand controller. The novel controller architecture is based on digital logic gates. A set of new methods to incrementally evolve the system is described. This includes several different variants of the fitness function being used. By applying the proposed schemes, the generalisation of the system is improved.

Jim Torresen

[1] David E. Goldberg,et al. Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[2] Jim Torresen,et al. Scalable evolvable hardware applied to road image recognition , 2000, Proceedings. The Second NASA/DoD Workshop on Evolvable Hardware.

[3] Jim Torresen,et al. Two-Step Incremental Evolution of a Prosthetic Hand Controller Based on Digital Logic Gates , 2001, ICES.

[4] Xin Yao,et al. Promises and challenges of evolvable hardware , 1996, IEEE Trans. Syst. Man Cybern. Part C.

[5] R N Scott,et al. Myoelectric prostheses: state of the art. , 1988, Journal of medical engineering & technology.

[6] Jim Tørresen,et al. A Divide-and-Conquer Approach to Evolvable Hardware , 1998, ICES.

[7] Isamu Kajitani,et al. An Evolvable Hardware Chip and Its Application as a Multi-Function Prosthetic Hand Controller , 1999, AAAI/IAAI.

[8] John Hallam,et al. Learning Complex Robot Behaviours by Evolutionary Computing with Task Decomposition , 1997, EWLR.

[9] Taro Nakamura,et al. Genetic Algorithm-Based Methodology for Pattern Recognition Hardware , 2000, ICES.

[10] Jordi Madrenas,et al. Evolvable Systems: From Biology to Hardware , 1996, Lecture Notes in Computer Science.