Task Allocation on Layered Multiagent Systems: When Evolutionary Many-Objective Optimization Meets Deep Q-Learning

This article is concerned with the multitask multiagent allocation problem via many-objective optimization for systems (MASs). First, a novel layered MAS model constructed to address that includes both original task simplification and allocation. In first layer of model, deep Q-learning method introduced simplify prioritization set. second modified shift-based density estimation (MSDE) put forward improve conventional strength Pareto evolutionary algorithm 2 (SPEA2) in order achieve on assignments. Then, an MSDE-SPEA2-based proposed tackle objectives including allocation, makespan, agent satisfaction, resource utilization, completion, waiting time. As compared existing methods, developed this exhibits outstanding feature assignment scheduling are carried out simultaneously. Finally, extensive experiments conducted to: 1) verify validity effectiveness two main algorithms 2) illustrate optimal solution efficient strategy under different scenarios.