A collaborative agent architecture with human-agent communication model
CAVE'12 Proceedings of the First international conference on Cognitive Agents for Virtual Environments
Hi-index | 0.00 |
Today, collaborative virtual environments consider intelligent software agents as essential part of the environment. In these environments, there is a need of agents that have the ability to learn user actions, which are gained from the experience of interactions with the users. This learning capability will predict the future actions of the user and provide better solutions to them. Collaborative Virtual Workspace (CVW) is a collaborative virtual environment where agents can be created and perform important tasks for the user. This paper presents an agent with the ability to monitor the user's actions and has the learning capability using reinforcement learning algorithm. To demonstrate the feasibility of this agent, it is implemented and demonstrated in FCVW an extension of CVW, an environment for collaboration and knowledge management.