Adaptive elevator group control with cameras

A new elevator group control system is proposed. The system learns dynamic traffic flows by analyzing passenger traffic without prespecified patterns. A genetic algorithm continuously generates dispatch functions according to changes in passenger traffic. By considering the status inside elevators, the directions of passenger movement, and the number of waiting passengers, the system occasionally allocates multiple elevators for a single hall call, which assists in reducing passengers' waiting time. Experimental results showed up to 25% improvement over a system without the above features.