My research interests include multi-agent systems, sophisticated negotiation and cooperation in multi-agent system, the design of intelligent agents, agent control and reasoning under uncertainty, learning in multi-agent system, and more generally, intelligent distributed systems and artificial intelligence. Here are some research projects I have been working on.
  • Sophisticated Negotiation in MultiAgent System
  •   Negotiation, an interactive communication among participants to resolve conflicts and get better solutions, is a very important issue for agents to effectively coordinate their behavior in Multi-Agent System(MAS).

    This work is motivated by following two questions. The first question is how an agent deals with multiple negotiation issues concurrently when these issues are related to each other. Another question is concerned with the negotiation in a complex organizational context. MultiAgent system consists of groups of loosely coupled agents that work together on tasks. The relationship between agents depends on their organizational role and could be any type of relationship between pure self-interested and totally cooperative. The agents have choices about with whom to collaborate, how to negotiate, what to charge for services, etc. The negotiation strategy is dependent on the relationship between the negotiating parties and the particular negotiation issue. The motivational qualities(MQ) framework developed earlier, is used to provide with the capability to reason about different objective goals hence the agent can evaluate a negotiation issue from the organizational objectives. The partial order schedule is exploited as a reasoning tool for the agent to handle the concurrent multiple linked negotiation issues and evaluate the flexibility and the feasibility in the negotiation. Also negotiation should be performed at different abstraction levels, rough commitments are formed at the upper level and then refined at the lower level to solve potential conflicts among different negotiation issues.
  • Cooperative, MultiStep Negotiation for Task Allocation
  •   A multi-dimensional, multistep negotiation mechanism is designed for task allocation among cooperative agents based on distributed search. This mechanism uses marginal utility gain and marginal utility cost to structure this search process, so as to find a solution that maximizes their combined utility. These two utility values together with temporal constraints summarize the agents' local information and reduce the communication load. This mechanism is anytime in character: by investing more time, the agents increase the likelihood of getting a better solution. A multiple attribute utility function is introduced into negotiations. This allows agents to negotiate over the multiple attributes of the commitment, which produces more options, making it more likely for agents to find a solution that increases the global utility. A set of protocols are constructed and the experimental result shows a phase transition phenomenon as the complexity of negotiation situation changes. A measure of negotiation complexity is developed that can be used by an agent to choose the appropriate protocol, allowing the agents to explicitly balance the gain from the negotiation and the resource usage of the negotiation.
  • Layered Architecture for Detailed Agent Coordination
  •   Multi-agent coordination is an important and complicated process. This project proposes a layered approach to coordination in which low-level domain independent coordination and scheduling modules deal with detailed temporal and resource constraints and high-level controllers focus on domain issues and domain state. A general agent architecture is designed based on this agent control model. The integration of the low-level controllers with high-level JIL process programming language is used to explore the model. The possibility of integrating with other frameworks such as domain planners and BDI-based controllers is also explored.
  • Resource-Bounded Information Gathering and Decision Making
  •   The vast amount of information available today on the World Wide Web (WWW) has great potential to improve the quality of decisions and the productivity of consumers. However, the WWW's large number of information sources and their different levels of accessibility, reliability and associated costs present human decision makers with a complex information gathering planning problem that is too difficult to solve without high-level filtering of information. In many cases, manual browsing through even a limited portion of the relevant information obtainable through advancing information retrieval (IR) and information extraction (IE) technologies is no longer effective. The time/quality/cost tradeoffs offered by the collection of information sources and the dynamic nature of the environment lead us to conclude that the user cannot (and should not) serve as the detailed controller of the information gathering (IG) process. Our solution to this problem is to integrate different AI technologies, namely scheduling, planning, text processing, and interpretation problem solving, into a single information gathering agent, BIG (resource-Bounded Information Gathering), that can take the role of the human information gatherer.

    I designed and implemented the RESUN Information Gathering Planner and the Execution Subsystem & Monitoring. Also I was responsible for the integration of the whole system and the experimental and the performance evaluation work.
  • Intelligent Home
  •   Intelligent environments are an interesting development and research application problem for multi-agent systems. The functional and spatial distribution of tasks naturally lends itself to a multi-agent model and the existence of shared resources creates interactions over which the agents must coordinate.

    The intelligent home project (IHome) at the UMASS multi-agent systems lab is an exploration in the application of multi-agent systems technology to the problem of managing an intelligent environment. We have implemented a sophisticated simulated home environment, populated it with distributed intelligent home-control agents (including simulated robots) that control appliances and negotiate over shared resources, and begun experimentation with different coordination protocols and agent adaptability/responsiveness to changing environmental conditions.

    I designed and implemented the Air Conditioner agent and the Heating agent which coordinate each other and also share resource with other agents.