PO-MADDPG：Multi-Agent Reinforcement Learning Algorithms
A multi-vehicle collaborative delivery algorithm based on multi-agent reinforcement learning is proposed, which applies the Multi-Agent Deep Reinforcement Learning algorithm MADDPG to the workshop delivery scenario.
Model Architecture Diagram：
Description： To address the problems of insufficient collaboration and overreliance on centralized control in the workshop delivery tasks, the Multi-Agent Deep Reinforcement Learning algorithm** MADDPG** is applied to the workshop delivery scenario with a focus on the vehicles as intelligent agent objects. Based on the partially observable environmental characteristics, the algorithm integrates material delivery information, vehicle carrying capacity, and surrounding vehicle information to optimize the multi-vehicle collaborative delivery task. The inputs to the algorithm are the number of agents and the scene size, and the output is a collaborative agent scheduling action.