With the rapid development of mobile energy storage systems （MESS），their importance in power system dynamic response

renewable energy integration

and emergency power supply has become increasingly prominent.However

challenges such as multi-objective resource allocation

path optimization under transportation-grid coupling constraints

and planning-scheduling coordination remain unresolved.Traditional methods struggle to address these issues due to computational complexity and insufficient dynamic adaptability.In contrast

artificial intelligence （AI）algorithms

leveraging data-driven technologies like reinforcement learning （RL）and graph neural networks （GNN）， have achieved breakthroughs in dynamic scheduling

collaborative optimization

and security-economy trade-offs.This paper systematically analyzes AI applications in MESS planning and scheduling

summarizes the advantages of deep reinforcement learning （DRL）in uncertainty decision-making and swarm intelligence （SI）algorithms in distributed coordination

and identifies research gaps in battery state of charge （SOC）modeling and cross-domain collaborative scheduling.Furthermore

an AI-empowered innovative framework for MESS planning and scheduling is proposed

offering theoretical foundations and practical pathways to enhance the resilience

economy

and low-carbon transition of power systems.