AWP-GAC: central-controlled actor-critic for multi-agent dynamic game environment

2021 ◽  
Author(s):  
Xinzhi li ◽  
Shengbo Dong ◽  
Xiangyang Cui
Keyword(s):  
Dynamic Game ◽  
Multi Agent ◽  
Game Environment

scholarly journals Modeling and Simulation of a Multi-agent Green Supply Chain Management System for Retailers

2020 ◽  
Vol 53 (4) ◽  
pp. 549-557
Author(s):  
Danfeng Wei
Keyword(s):  
Supply Chain ◽  
Management System ◽  
Dynamic Game ◽  
Evaluation Index System ◽  
Agent Technology ◽  
Green Supply Chain ◽  
Multiple Agents ◽  
Multi Agent ◽  
Supply Chain Management System ◽  
Hierarchical Evaluation

In recent years, the agent technology has been successfully applied in supply chains, thanks to its excellent interactivity, proactivity, and autonomy. However, the existing research on multi-agent green supply chain (GSC) stops on the strategic and tactical levels, failing to implement the relevant supply chain models. To overcome the limitation, this paper designs a multi-agent GSC management system for retailers, with the aim to obtain scientific collaboration strategies among multiple agents and to make effective logistics decisions for the supply chain. Firstly, a 3-layer hierarchical evaluation index system (EIS) was established under the framework of the multi-agent GSC management system. Next, the authors modeled the supply-demand relationship and dynamic game of multi-agent GSC, and discussed the cooperation and negotiation models among multiple agents. Experimental results show that the proposed multi-agent GSC management model is highly effective. The research findings provide a reference for the application of multi-agent technology in other types of supply chain enterprises.

scholarly journals Development of Intelligent Multi-Agent in the Game Environment

2015 ◽  
Vol 16 (6) ◽  
pp. 69-78
Author(s):  
DongMin Kim ◽  
JinWoo Choi ◽  
ChongWoo Woo
Keyword(s):  
Multi Agent ◽  
Game Environment

scholarly journals Patrolling AI Systems in Video Games

2019 ◽  
Vol 8 (4) ◽  
pp. 2479-2485
Keyword(s):  
Artificial Intelligence ◽  
Video Games ◽  
Real Time ◽  
Dynamic Game ◽  
Common Element ◽  
Path Finding ◽  
Game Engine ◽  
Adventure Games ◽  
Game Environment ◽  
Moving Obstacle

Artificial Intelligence is a common element in digital video games and it is one of the most essential component in modern games. Modern games are populated with Non-Player Characters (AI Characters) that performs different activities. Realism is dominating in games and AI behavio rs are expected to be more realistic in games. Games that has poor unrealistic AI elements are facing heavy criticism from the player bases. Further, modern games are highly dynamic. Classical games had static environment with less or no changes in it. Such environments made implementation of AI easy and simple. In modern games, Progressive terrain generation and other such content generation methods increases the complexities of building an efficient AI for games that has many changes in real time. One of the most common AI in games is Patrolling AI especially in Shooter and Adventure Games. Patroling AI involves path finding and obstacle attack or defense. RRT algorithm and its variants are highly successful Probabilistic Determination AI that produced effective results in real time robotic movement. In order to build efficient Patrolling AI for games, a real time RRT* variant called RT-RRT* algorithm was employed. The algorithm is flexible enough to add various behaviors in addition to path finding which makes it more suitable for games. The algorithm takes samples from the environment and construct the efficient path. Also the algorithm inspect the environment in run time to ensure that no moving obstacle blocks the path. In such case, it rewires and create a new path. In order to manage the dynamic obstacles, a Real Time Obstacle Handling Algorithm was designed and employed in a dynamic game environment. The algorithms inputs the obstacles types and parameters. On identifying the obstacle approaching the AI in terrain, it tells the agent to perform the needed actions. The simulations was carried out using Unity Game Engine. The model proposed helped to create efficient patrolling AI that handle two major aspects of patrolling which is Path finding and Obstacle handling. The model will be highly suitable for dynamic game environments with lots of uncertainty and emergence.

Research on the Behavior of Intelligent Role in Computer Games Based on Behavior Tree

2014 ◽  
Vol 509 ◽  
pp. 165-169 ◽  
Author(s):  
Li Xia Ji ◽  
Jian Hong Ma
Keyword(s):  
Computer Games ◽  
Dynamic Game ◽  
Role Behavior ◽  
Complex Processes ◽  
Behavior Tree ◽  
Finite State ◽  
Game Environment ◽  
Simple Logic ◽  
And Control ◽  
Tree Method

Design of intelligent role behavior about computer games is a challenge in the artificial intelligence (AI) field. Finite state machine (FSM) is often used to design the behavior of games. The limitation of FSM is using simple logic to design complex intelligence role behavior, and it cant produce reasonable interactive role behavior in dynamic game environment. This paper studied and proposed a generic behavior tree method. This method can achieve AI framework effectively. It is a generic and maintainable tool for complex processes management. Behavior tree is hierarchical, and all of its layers are parallel. Through conditions and control node of behavior tree, we can use the graphical control various branching behavior clearly, and thus respond more complex intelligent role behavior.

scholarly journals The Convergence of a Cooperation Markov Decision Process System

Entropy ◽  
2020 ◽  
Vol 22 (9) ◽  
pp. 955
Author(s):  
Xiaoling Mo ◽  
Daoyun Xu ◽  
Zufeng Fu
Keyword(s):  
Markov Decision Process ◽  
Decision Process ◽  
Optimal Strategy ◽  
Value Function ◽  
Single Agent ◽  
Two Agents ◽  
Markov Decision ◽  
Process System ◽  
Multi Agent ◽  
Game Environment

In a general Markov decision progress system, only one agent’s learning evolution is considered. However, considering the learning evolution of a single agent in many problems has some limitations, more and more applications involve multi-agent. There are two types of cooperation, game environment among multi-agent. Therefore, this paper introduces a Cooperation Markov Decision Process (CMDP) system with two agents, which is suitable for the learning evolution of cooperative decision between two agents. It is further found that the value function in the CMDP system also converges in the end, and the convergence value is independent of the choice of the value of the initial value function. This paper presents an algorithm for finding the optimal strategy pair (πk0,πk1) in the CMDP system, whose fundamental task is to find an optimal strategy pair and form an evolutionary system CMDP(πk0,πk1). Finally, an example is given to support the theoretical results.

scholarly journals Double Layer Dynamic Game Bidding Mechanism Based on Multi-Agent Technology for Virtual Power Plant and Internal Distributed Energy Resource

Energies ◽  
2018 ◽  
Vol 11 (11) ◽  
pp. 3072 ◽  
Author(s):  
Yajing Gao ◽  
Xiaojie Zhou ◽  
Jiafeng Ren ◽  
Xiuna Wang ◽  
Dongwei Li
Keyword(s):  
Power Plant ◽  
Double Layer ◽  
Dynamic Game ◽  
Internal Market ◽  
Agent Technology ◽  
Bidding Strategy ◽  
Virtual Power ◽  
Distributed Energy ◽  
Virtual Power Plant ◽  
Multi Agent

As renewable energies become the main direction of global energy development in the future, Virtual Power Plant (VPP) becomes a regional multi-energy aggregation model for large-scale integration of distributed generation into the power grid. It also provides an important way for distributed energy resources (DER) to participate in electricity market transactions. Firstly, the basic concept of VPP is outlined, and various uncertainties within VPP are modeled. Secondly, using multi-agent technology and Stackelberg dynamic game theory, a double-layer nested dynamic game bidding model including VPP and its internal DERs is designed. The lower layer is a bidding game for VPP internal market including DER. VPP is the leader and each DER is a subagent that acts as a follower to maximize its profit. Each subagent uses the particle swarm algorithm (PSA) to determine the optimal offer coefficient, and VPP carries out internal market clearing with the minimum variance of unit profit according to the quoting results. Then, the subagents renew the game to update the bidding strategy based on the outcomes of the external and internal markets. The upper layer is the external market bidding game. The trading center (TC) is the leader and VPP is the agent and the follower. The game is played with the goal of maximum self-interest. The agent uses genetic algorithms to determine the optimal bid strategy, and the TC carries out market clearance with the goal of maximizing social benefits according to the quotation results. Each agent renews the game to update the bidding strategy based on the clearing result and the reporting of the subagents. The dynamic game is repeated until the optimal equilibrium solution is obtained. Finally, the effectiveness of the model is verified by taking the IEEE30-bus system as an example.

scholarly journals Coordinated Development Based Grid-Source-load Collaborative Planning Method of Uncertainty and Multi-agent Game

2021 ◽  
Vol 233 ◽  
pp. 01115
Author(s):  
He Lei ◽  
Si-xuan Zhou ◽  
Yin-xiang Wang ◽  
Yun-fei Zheng ◽  
Lin Jing ◽  
...  
Keyword(s):  
Robust Optimization ◽  
Power Grid ◽  
Collaborative Planning ◽  
Dynamic Game ◽  
Planning Method ◽  
Optimal Decision ◽  
Game Player ◽  
Multi Agent ◽  
Deep Integration ◽  
Virtual Game

When planning the power grid, it is necessary to obtain the optimal decision scheme according to the market behavior of different stakeholders. In this paper, the virtual game player "nature" is introduced to realize the deep integration of game theory and robust optimization, and a source network load collaborative planning method considering uncertainty and multi-agent game is proposed. Firstly, the planning decision-making models of different stakeholders of DG investment operators, power grid investment operators and power users are constructed respectively; then, the static game behavior between distributed generation (DG) investment operators and power grid investment operators is analyzed according to the transmission relationship of the three; at the same time, robust optimization is used to deal with DG. In this paper, we introduce the virtual game player "nature" to study the dynamic game behavior between the virtual game player and the power grid investment operator. On this basis, the dynamic static joint game planning model is proposed.

Sign in / Sign up

Export Citation Format

Share Document