Micro-genetic algorithms in intelligent traffic signal control

This paper presented a micro-GA based algorithm to optimize dynamical traffic signal control systems operating in oversaturated conditions. The objective of the control algorithm is to find, for a given control period, the near-optimal control trajectory (green splits an offsets) for a series of closely spaced traffic signals along an oversaturated arterial such that system throughput is maximized. The problem was solved using a micro-Genetic Algorithm (micro-GA). GAs were used because of their robustness, adaptive capabilities, and ability to overcome combinatorial explosions typical of highly dimensional problems like the one at hand. Despite the vast size of the solution set, the GA was able to converge to a near-optimal solution in very short time. The results show that the control algorithm provides efficient traffic control such that undesirable conditions such as queue build-up and spill-back are prevented. Intelligent and adaptive capabilities such as these would be of critical value in an intelligent transportation systems (ITS) environment, particularly during oversaturated conditions.