论文部分内容阅读
针对粒子群算法对约束条件的优化处理问题,提出一种具有自适应度双群体粒子群优化算法,该算法将目标函数与约束条件分别考虑,形成2种群体以不同目标为前提同时向最优解进化;并分别对2种群体的适应度引入自适应权重系数与相应调整策略,基于并非所有非可行个体均劣于可行个体概念,动态地调整其适应度以保证部分非可行个体向可行域进化。将其应用于组群机器人队形控制中,链型结构(纵队)队形仿真结果表明了该算法的有效性。该粒子群算法为实际应用中约束优化问题的求解提供了新的途径。
In order to solve the optimization problem of particle swarm optimization (PSO), a dual-particle swarm optimization algorithm with adaptive degree is proposed. The algorithm considers the objective function and constraints respectively, and forms two kinds of groups, Based on the notion of feasible individuals, we adaptively adjust their fitness to ensure that some of the non-viable individuals to the feasible region evolution. Applying this method to the formation control of a group robot, the simulation result of the chain structure (column) formation shows the effectiveness of the algorithm. The PSO algorithm provides a new way to solve the constrained optimization problem in practice.