Mixed-traffic signal timing at signalized intersection based on adaptive co-evolutionary algorithm

In order to solve the signal timing of the urban single-point intersection under the mixed-traffic, the shortest average delay were taken as objective, and the intersection geometry shape, channelization status and prohibition of traffic flow conflicts were taken as restrictions. A new global search algorithm combining adaptive genetic algorithm with co-evolutionary algorithm was put forward to solve the optimization problem, and simulation test with the actual data of Tiyuchang-Jianguo intersection indicates that the proposed algorithm overcomes the shortcomings such as local convergence, compared with traditional genetic algorithm; and it is feasible and effective for mixed-traffic signal timing at signalized intersection.