A smart factory-oriented edge collaborative computing task offloading method and system

Abstract

The application relates to the technical field of industrial Internet of Things and edge computing, and discloses a kind of edge-end collaborative computing task unloading method and system for intelligent factory, which comprises the following steps: establishing a task unloading decision mechanism of multiple devices and multiple edge servers, assigning a computing node to each task, determining task delay, and constructing a node expertise matrix; an optimization model is established; the optimization problem of the optimization model is formalized as a Markov decision process; a deep reinforcement learning algorithm is used to solve the Markov decision process, and the intelligent workshop environment is interacted to store the generated experience data into the experience cache area; the network parameters of the deep reinforcement learning algorithm and the node expertise matrix corresponding to the edge server are synchronously updated using the data pairs of the experience cache area; until the deep reinforcement learning algorithm converges, and the policy network and the node expertise matrix are output. The optimal task allocation decision is dynamically generated, and the completion deviation of the task is effectively reduced.

Images