Both simple and hybrid genetic algorithms encounter difficulties when presented with a function which has multiple values. Similarly, changing environments or functions which change rapidly present other problems. This paper presents an algorithm that is capable of coping with both of these scenarios: it can accommodate multiple solutions simultaneously and can track changes in optima efficiently. The proposed B-cell algorithm is inspired by the natural immune system, which itself displays similar capabilities of tracking multiple, moving targets in the form of infectious agents. This paper employs two nonlinear mappings which display chaotic behaviour to demonstrate the effectiveness of the B-cell algorithm in tracking multiple, moving targets. A number of experiments are conducted and results reported from the B-cell algorithm and standard hybrid genetic algorithm approaches. These results show the benefit of the B-cell algorithm approach when compared against these heuristic approaches.
|Number of pages||7|
|Publication status||Published - 2003|
|Event||2003 Congress on Evolutionary Computation, CEC 2003 - Canberra, ACT, Australia|
Duration: 08 Dec 2003 → 12 Dec 2003
|Conference||2003 Congress on Evolutionary Computation, CEC 2003|
|Period||08 Dec 2003 → 12 Dec 2003|