An evolutionary program for a class of continuous optimal control problems

We describe an evolutionary programming technique for finding the solution to classical continuous optimal control problems, with or without free final time. We define a set of highly abstracted genetic operators to manipulate the gene pool in a `blind´ but meaningful way. We apply the algorithm to Luus´s (1990) highly non-linear model of oil-shale pyrolysis, and to an iso-thermal continuous stirred tank reactor with 3 independent control variables