Self-adaptive multi-objective optimization method design based on agent reinforcement learning for elevator group control systems

This paper study the multi-objective optimization problem of elevator group control systems by using the Markov Decision Process model. Define the Agent to be the leaner and decision-maker of the MDP model. And then using reinforcement learning Algorithm combined with generic method defines the elements of this model. Moreover we use SARSA(λ) value iteration algorithm which was selected to iterative estimation the utility function combined with tile coding function approximation to design an optimization algorithm, and then prove that the solution of this algorithm will converges to a bounded domain which is given in this paper. The effect for dynamic optimization objective function of proposed approach was validated by virtual simulation environment of elevator group control systems.