An improved Multi‐Agent Proximal Policy Optimization (MAPPO)‐based scheduling framework is proposed to enable distributed and cooperative decision‐making under partial observability and achieves lower task latency and energy consumption, improved load balancing, and more stable convergence compared with state‐of‐the‐art scheduling methods.
Abstract
With the rapid proliferation of IoT devices and latency‐sensitive applications such as autonomous driving and immersive reality, edge computing has become essential for meeting stringent delay and bandwidth requirements. However, the increasing volume of computation‐intensive tasks, together with resource heterogeneity and environmental dynamics, poses significant challenges to efficient task scheduling in edge environments. To address these issues, this paper formulates a multi‐objective task scheduling problem for mobile edge computing systems with densely deployed base stations as a Partially Observable Markov Decision Process (POMDP). An improved Multi‐Agent Proximal Policy Optimization (MAPPO)‐based scheduling framework is proposed to enable distributed and cooperative decision‐making under partial observability. To further improve inter‐agent coordination and global optimization capability, cooperative co‐evolutionary mechanisms are incorporated, while neuroevolution is employed to jointly optimize the policy network architecture and parameters. Extensive simulations demonstrate that the proposed approach achieves lower task latency and energy consumption, improved load balancing, and more stable convergence compared with state‐of‐the‐art scheduling methods.
To address the requirement of latency‐sensitive intelligent transportation applications, vehicular fog computing has proven to be a promising paradigm to push the computation towards the vehicles. However, efficient task offloading and scheduling is a challenging optimization problem because of the highly dynamic natur...
Alaguvathana Paravel, Dhurgadevi Muthusamy· International Journal of Net...· 0 citations
This paper proposes a heterogeneous multi-agent proximal policy optimization (MAPPO)-based framework where both user devices and UAVs act as heterogeneous agents and utilizes a centralized training and decentralized execution (CTDE) paradigm to enable collaborative strategies between computing requesters and providers.
Ming Cheng, Canlin Zhu, Jian-Hang Tang et al.· Journal of King Saud Univers...· 0 citations
This paper innovatively proposes a scheduling framework called SPARCD that integrates lightweight position prediction mechanisms, action masking mechanisms, and state filling techniques to construct a multi-objective scheduling model that jointly optimizes task latency, energy consumption, and reliability.
Li-Qiong Chen, Liu-Bing Xiao, Xi Chen et al.· Cluster Computing· 0 citations
This work proposes a multi-agent reinforcement learning (MARL) approach for queue-level scheduling in TSN-enabled VEC and employs multi-agent proximal policy optimization (MAPPO) to enable coordinated yet autonomous scheduling decisions.
B. A. C. Pereira, Marcos Carvalho, Fatih Temiz et al.· 0 citations
Rooted in Stackelberg game theory and reinforced by the Heterogeneous Earliest Finish Time heuristic, DLFS enables dynamic, priority-aware task coordination across fog and cloud layers, and effectively reconciles competing objectives while respecting varying deadline and budget constraints.
Marzieh Soleimani, A. Deldari· Computing· 0 citations
This study proposes a multi-objective optimization-based framework for task allocation and path planning to address the challenges faced by multi-robot systems in transport-oriented task environments. The framework considers robot capability heterogeneity and load capacity, aiming to minimize task execution time and ov...
S.-H. Lo, R. Chen· Automation· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.