Page 115 - Innovations in Intelligent Machines

UAV Path Planning Using Evolutionary Algorithms 105

Fig. 9. The fourth test case for the coordinated UAV path planning

Fig. 10. Convergence histories for the first test case, with and without the use of
the RBFN assistance

surrogate model, the convergence histories for the four test cases of the DE
algorithm (with and without the RBFN assistance) are presented in Fig. 10 to
13. As it can be observed, the adoption of the approximation model resulted
in a considerable reduction in the number of exact evaluations for a specific
fitness value. This reduction, for all cases considered, reached or exceeded
80% of the number of exact evaluations with respect to the case without