Experiments on a Prey Predators System

摘要

The paper describes a prey-predators system devoted to perform experiments on concurrent complex environment. The problem has be treated as an optimization problem. The prey goal is to escape from the predators reaching its lair, while predators want to capture the prey. At the end of the 19th century, Pareto found an optimal solutions for decision problems regarding more than one criterion at the same time. In most cases this ‘Pareto-set’ cannot be determined analytically or the computation time could be exponential. In such cases, evolutionary Algorithms (EA) are powerful optimization tools capable of finding optimal solutions of multi-modal problems. Here, both prey and predators learn into an unknown environment by means of genetic algorithms (GA) with memory. A set of trajectories, generated by a GA, are able to build a description of the external scene driving a predators to a prey and the prey to the lair. The prey-predator optimal strategies are based on field of forces. This approach could be applied to the autonomous robot navigation in risky or inaccessible environments (monitoring of atomic power plants, exploration of sea bottom, and space missions).