An iterative evolution algorithm for selecting the logic rules and tuning the membership functions for a signalized intersection with genetic fuzzy logic controller (GFLC) is proposed. The GFLC model employs traffic flow and queue length as state variables, extension of green time as control variable and total vehicle delays estimated by fluid approximation method as performance measurement criterion. Validations from an experimental example and a field case show that our GFLC model can perform almost the same as the optimal multiple timing plan and far superior to the optimal single, Webster, and current timing plans. As traffic flows randomly vary by 10%, 30% and 50%, the GFLC model can even perform better than the optimal multiple timing plan. The results suggest that our GFLC model is effective, robust and applicable for adaptive traffic signal control.