基于GA的交叉路口自适应模糊控制器优化设计

Optimized Designation of Adaptive Fuzzy Traffic Controller Based on Genetic Algorithm

经典交叉路口交通信号模糊逻辑控制器的模糊规则是预先人为确定的,不可避免地带有主观因素,且不能随着交通流的变化而动态改变,不能保证模糊逻辑控制系统最优的要求。本文提出适用于交通信号模糊逻辑控制的改进GA算法,设计适用于模糊规则编码的带区间范围限制的十进制编码方案,并实现受限分布一致交叉和变异算子,改进经典赌轮选择法,避免病态个体产生,加快收敛速度,可以自适应地动态确定最优模糊规则。MATLAB仿真结果表明该方法的有效性。

Conventional fuzzy traffic signal controllers use man-made control rules.The this approach does not guarantee the optimal solution to design dynamic fuzzy control system.In this paper,the near optimal rules of traffic fuzzy controller were automatically determined by the enhanced genetic algorithm method which is refined for fuzzy traffic signal control.A new decimal coding schema was designed to implement the unified coding of fuzzy rules.An improved selection method,a constrained and stepped crossover and mutation operator are presented to decrease the sick-items and speed up the convergence.The effectiveness of the method was shown through MATLAB simulation.