Simulation-based algorithms for the optimization of sensor deployment
Résumé
Two simulation-based algorithms are presented, that have been successfully applied to an industrial optimization problem. These two algorithms have different and complementary features. One is fast, and sequential: it proceeds by running a population of targets and by dropping and activating a new sensor (or re-activating a sensor already available) where and when this action seems appropriate. The other is slow, iterative, and non-sequential: it proceeds by updating a population of deployment plans with guaranteed and increasing criterion value at each iteration, and for each given deployment plan, there is a population of targets running to evaluate the criterion. Finally, the two algorithms can cooperate in many different ways, to try and get the best of both approaches. A simple and efficient way is to use the deployment plans provided by the sequential algorithm as the initial population for the iterative algorithm.