Diffusion Reinforcement Learning Based Online 3D Bin Packing Spatial Strategy Optimization
For logistics and warehousing, this work improves packing efficiency in complex online scenarios, though the improvement is incremental over existing DRL methods.
The paper tackles the online 3D bin packing problem using deep reinforcement learning. The proposed diffusion reinforcement learning algorithm significantly improves the average number of packed items compared to state-of-the-art DRL methods.
The online 3D bin packing problem is important in logistics, warehousing and intelligent manufacturing, with solutions shifting to deep reinforcement learning (DRL) which faces challenges like low sample efficiency. This paper proposes a diffusion reinforcement learning-based algorithm, using a Markov decision chain for packing modeling, height map-based state representation and a diffusion model-based actor network. Experiments show it significantly improves the average number of packed items compared to state-of-the-art DRL methods, with excellent application potential in complex online scenarios.