带自适应整定参数的机器人达尔文粒子群优化算法

Robot Darwinian Particle Swarm Optimization with Self-adaptive Tuning Parameters

对基数庞大的机器人群族引入达尔文粒子群优化算法(DPSO)。该算法将自然选择应用到粒子群算法中,对整个机器人群族进行动态分割,根据上下文评价指标配合机器人行为对机器人的行为进行预测,提高了机器人群族运动的最优逃脱方案成功率。仿真试验表明,通过对该算法的输入参数进行自适应整定,可以改进系统的收敛率,增加通信的约束,使整个机器人群族在未来更大的范围内有效驱动数量更大的无线机器人群族。

The Darwinian particle swarm optimization （ DPSO ） is introduced in the robots swarm with tremendous cardinality. The algorithm applies natural choice in particle swarm algorithm, dynamically divides the entire robots swarm, and predicts the behavior of robots according to the context evaluation indicator with robot＇ s behavior, to increase the optimal escape rate of the motion of robots swarm. The simulation tests show that through adaptive tuning of the input parameters of the algorithm, the convergence rate of the system can be improved, the communication constrain is increased, which lead to larger wireless robots swarm can be efficiently driven by entire robots swarm in larger scope in the future.