• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.11-19
• /
• 2024

Multi-agent systems can be utilized in various real-world cooperative environments such as battlefield engagements and unmanned transport vehicles. In the context of battlefield engagements, where dense reward design faces challenges due to limited domain knowledge, it is crucial to consider situations that are learned through explicit sparse rewards. This paper explores the collaborative potential among allied agents in a battlefield scenario. Utilizing the Multi-Robot Warehouse Environment(RWARE) as a sparse reward environment, we define analogous problems and establish evaluation criteria. Constructing a learning environment with the QMIX algorithm from the reinforcement learning library Ray RLlib, we enhance the Agent Network of QMIX and integrate Random Network Distillation(RND). This enables the extraction of patterns and temporal features from partial observations of agents, confirming the potential for improving the acquisition of sparse reward experiences through intrinsic rewards.

• Journal of the Korea Society for Simulation
• /
• v.29 no.2
• /
• pp.105-117
• /
• 2020

Analysis of combat effectiveness is required to consider the concept of tactical cooperative engagement between manned-unmanned weapon systems, in order to predict the required operational capabilities of future weapon systems that meets the concept of 'effect-based synchronized operations.' However, analytical methods such as mathematical and statistical models make it difficult to analyze the effects of complex systems under nonlinear warfare. In this paper, we propose a combat simulation model that can simulate the concept of cooperative engagement between manned-unmanned combat entities based on wireless communications. First, we model unmanned combat entities, e.g., unmanned ground vehicles and drones, and manned combat entities, e.g., combatants and artillery, considering the capabilities required by the future ground system. We also simulate tactical behavior in which all entities perform their mission while sharing battlefield situation information through wireless communications. Finally we explore the feasibility of the proposed model by analyzing combat effectiveness such as target acquisition rate, remote control success rate, reconnaissance lead time, survival rate, and enemy's loss rate under a small-unit armor reconnaissance scenario. The proposed model is expected to be used in war-game combat experiments as well as analysis of the effects of manned-unmanned ground weapons.