Adaptive cooperative ant colony optimization algorithm based on reinforcement incentive and its application

Abstract

Ant colony algorithm has great advantages in solving some NP complete problems, but it also has some problems such as slow search speed, low convergence accuracy and easy to fall into local optimum. In order to balance the contradiction between the convergence accuracy and the convergence speed of ant colony algorithm, this paper first proposes an ant colony algorithm (RIACO) based on the reinforcement excitation theory of Burrus Frederic Skinner. In this algorithm, pheromone is stimulated and its volatilization coefficient is adjusted adaptively according to the iteration times, thus the speed of ant colony search is accelerated. Secondly, based on the characteristics of real ant colony classification and division of labor, this paper proposes an ant colony algorithm based on labor division and cooperation (LCACO). The algorithm divides the ant colony into two different types of ant colony for information exchange and improves the state transition probability formula, so that the two ant colonies can search the optimal path cooperatively, so as to improve the precision of ant colony search. Finally, combining the two improved ant colony algorithms, this paper proposes an adaptive cooperative ant colony optimization algorithm based on reinforcement incentive (SMCAACO). A multi constrained vehicle routing problem (MCVRP) is compared with the classical tabu search algorithm (TS), variable neighborhood search algorithm (VNS) and basic ant colony algorithm (ACO). The experimental results show that, in solving the mcvrp problem, the algorithm proposed in this paper not only has a good performance in the solution results, but also achieves a good balance between the convergence speed and the convergence accuracy.