Research On Active Perception And Autonomous Decision For Smart Warehouse Logistics

As an important part of the national strategy of intelligent manufacturing,the construction of smart warehouse logistics has become a key link in the innovation driven development of manufacturing supply chains.In view of the challenges of complex operation processes,high pressure on heterogeneous data processing,unreasonable allocation of storage locations,and low efficiency in picking and distribution in modern warehouse logistics,this paper studies the issues of massive data perception and monitoring mechanisms,autonomous allocation of storage locations,and joint scheduling of picking and distribution orders for smart warehouse logistics.Firstly,in response to the demands for the complexity of smart warehouse logistics and slow response speed of data processing,an active perception and collaborative computing framework based on the fusion of "human-cyber-physical" is proposed.The data collection and real-time monitoring of warehouse operations are realized based on RFID technology.From a closed-loop perspective of "planning-monitoring-perception-decision",the collaborative operation logic of "human-cyber-physical" in warehouse scenarios such as inbound,storage,outbound,and delivery are explicated.Secondly,in response to the optimization problem of smart warehouse location allocation,considering factors such as diverse optimization objectives and dynamic changes in assignable locations,the inbound order information and the monitoring diagram model of the inbound operation are analyzed.With the goal of minimizing operation time and improving shelf stability,an autonomous allocation model for inbound storage locations in the scenario of parallel operation of stackers is established.An improved mayfly algorithm is designed to solve the above model.The case analysis results show that this improved mayfly algorithm can obtain satisfactory solutions compared to other algorithms in different order scales,and can effectively reduce the operation time of storage.Then,in response to the problem that the picking and distribution operations of warehouse orders are independent of each other and difficult to achieve the overall optimum efficiency,the "1+N+M" model of outbound operation logic is analyzed.The picking order of outbound orders is combined with the customer information required for service to establish a mathematical model for joint scheduling of picking and distribution.A three-stage algorithm is used to optimize the delivery route and adjust the picking order of goods in the outbound orders,thereby expanding the collaborative operation level of smart warehouse logistics picking and distribution.Finally,based on the aforementioned research,a prototype system for smart warehouse active perception and autonomous decision is designed and developed using Python language.Through case validation,the feasibility of smart warehouse perception interaction,location allocation,picking and distribution joint scheduling models and algorithms are demonstrated.