Advanced Agent - Robotics Technology Lab - The Robotics Institute Carnegie Mellon University
Home/Advanced Agent – Robotics Technology Lab
hero image
Advanced Agent – Robotics Technology Lab
Here, researchers explore multi-agent planning and scheduling, multi-agent learning, multi-agent negotiation, and decision support for human teams

Agent Interoperability: We are designing reusable agent types, architectures, and languages to address the problem of interagent communication, coordination and negotiation.

Electronic Commerce: We are addressing the need for agent-based negotiation and decision techniques that can be applied directly to the challenges facing the electronic marketplace.

Information Gathering: The rapid growth of electronic information threatens to overwhelm human attention, raising new challenges for information retrieval (IR) technology. We have developed several technologies that address the need to make information useful by the automatic discovery of relevant, yet previously unknown information, the tracking of important topics, and the customization of information delivery to human users, amongst other technologies.

Middle Agents: We are implementing systems of agent brokers and liaisons to manage cross-agent communication and task sharing.

Multi-Agent Systems: We are solving increasingly complex problems across fluid and spatially distributed environments by implementing systems of heterogeneous multi-agent teams that decompose, delegate, and negotiate tasks.

Multi-Agent Learning: We are exploring agent learning mechanisms from agent-agent and agent-environment perspectives.

Multi-Agent System Interoperability: We investigate the design of middle agents for interoperability across multi-agent system boundaries. This agent architecture supports scalability by providing highly cost-effective means by which agents from two very different multi-agent system architectures can share capabilities, services, and information.

Security: We have researched and developed technologies for security in agent communities and security in the special domain of ecommerce transactions.

Semantic Web Services: Today, Web Services are discovered and invoked manually by human users, limiting the ability to take advantage of opportunities that may exist. Our research efforts are aimed at overcoming these limitations, allowing Web Services to discover and interact with each other autonomously, thus reducing the need for human mediation. This research will help us to realize the vision for the next Web revolution, the transformation of the Web from a web of static pages to a population of autonomous Web Services.

Displaying 207 Publications


We are developing DAML as part of the Semantic Web initiative.

Urban Search and Rescue

We are developing Hybrid Teams of Autonomous Agents: Cyber Agents, Robots and People (CARPs) to address the challenges of urban search and rescue.


A personal agent for World-Wide Web browsing that enhances searches and learns user interests.

Share This Story!