Resource Allocation for Metaverse Experience Optimization: A Multi-Objective Multi-Agent Evolutionary Reinforcement Learning Approach

In the Metaverse, real-time, concurrent services such as virtual classrooms and immersive gaming require local graphic rendering to maintain low latency. However, the limited processing power and battery capacity of user devices make it challenging to balance Quality of Experience (QoE) and terminal energy consumption. In this paper, we investigate a multi-objective optimization problem (MOP) regarding power control and rendering capacity allocation by formulating it as a multi-objective optimization problem. This problem aims to minimize energy consumption while maximizing Meta-Immersion (MI), a metric that integrates objective network performance with subjective user perception. To solve this problem, we propose a Multi-Objective Multi-Agent Evolutionary Reinforcement Learning with User-Object-Attention (M2ERL-UOA) algorithm. The algorithm employs a prediction-driven evolutionary learning mechanism for multi-agents, coupled with optimized rendering capacity decisions for virtual objects. The algorithm can yield a superior Pareto front that attains the Nash equilibrium. Simulation results demonstrate that the proposed algorithm can generate Pareto fronts, effectively adapts to dynamic user preferences, and significantly reduces decision-making time compared to several benchmarks.

I. Introduction

The Metaverse is characterized by its immersive nature, allowing seamless integration between physical and virtual experiences, offering users the ability to live, work, and play in a digital environment [1]. For example, hands-on training with immersive simulations enables workers to refine their skills within a risk-free virtual realm [2]. Using virtual classrooms would allow students and teachers to interact in 3D virtual spaces for lectures, discussions, and collaborative activities [3]. Compared to the passive experience of live video streaming, the Metaverse is distinguished by its high level of interactivity. Real-time immersive Metaverse services place rigorous requirements on hardware and software to provide low delay and fast rendering performance. In terms of hardware, lightweight, low-power head-mounted displays (HMDs) and physical assistive devices are required to use high-resolution images for a long time without frequent charging or battery replacement [4]. In terms of software, rising delay adversely affects the user experience, potentially leading to dizziness and motion sickness due to sensory confusion [5]. Due to the limitations in computing power and battery capacity of HMD devices, balancing terminal energy consumption while ensuring immersive user experience and achieving high-quality, low-latency rendering on low-power devices has become an urgent challenge to address.