The Fifteenth Annual AAAI Robot Competition and Exhibition was held at the Twenty-First National Conference on Artificial Intelligence in Boston, Massachusetts, in July 2006. This article describes the events that were held at the conference, including the Scavenger Hunt, Human Robot Interaction, and Robot Exhibition.

Customer service is defined as the ability of a company to afford the service requestor with the expressed need. Due to the increasing importance of service offerings as a revenue source and increasing competition among service providers, it is important for companies to optimize both the customer experience as well as the associated cost of providing the service. For more complex interactions with higher value, mixed-initiative systems provide an avenue that gives a good balance between the two goals. This article describes a mixed-initiative system that was created to improve customer support for problems customers encountered with their appliances. The tool helped call takers solve customers' problems by suggesting questions aiding the diagnosis of these problems. The mixed-initiative system improved the correctness of the diagnostic process, the speed of the process, and user satisfaction. The tool has been in use since 1999 and has provided more than $50 million in financial benefits by increasing the percentage of questions that could be answered without sending a field service technician to the customers' homes. Another mixed-initiative tool, for answering e-mail from customers, was created in 2000.

The MAPGEN system represents a successful mission infusion of mixed-initiative planning technology. MAPGEN was deployed as a mission-critical component of the ground operations system for the Mars Exploration Rover mission. Each day, the ground-planning personnel employ MAPGEN to collaboratively plan the activities of the "Spirit and "Opportunity rovers, with the objective of achieving as much science as possible while ensuring rover safety and keeping within the limitations of the rovers' resources. The Mars Exploration Rover mission has now been operating for more than two years, and MAPGEN continues to be employed for activity plan generation for the Spirit and Opportunity rovers. During the multiyear deployment effort and subsequent mission operations experience, we have learned valuable lessons regarding application of mixed-initiative planning technology to mission operations. These lessons have spawned new research in mixed-initiative planning and have influenced the design of a new ground operations system, called M-SLICE, that is baselined for the Mars Science Laboratory mission. In this article, we discuss the mixed-initiative aspects of the MAPGEN system, focusing on the task, control, and awareness issues.

Mixed-initiative planning systems attempt to integrate human and AI planners so that the synthesis results in high-quality plans. In the AI community, the dominant model of planning is search. In state-space planning, search consists of backward and forward chaining through the effects and preconditions of operator representations. Although search is an acceptable mechanism to use in performing automated planning, we present an alternative model to present to the user at the interface of a mixed-initiative planning assistant. That is, we propose to model planning as a goal-manipulation task. Here planning involves moving goals through a hyperspace in order to reach equilibrium between available resources and the constraints of a dynamic environment. The users can establish and "steer" goals through a visual representation of the planning domain. They can associate resources with particular goals and shift goals along various dimensions in response to changing conditions as well as change the structure of previous plans. Users need not know the details of the underlying technology, even when search is used within. This article empirically examines user performance under both the search and the goal-manipulation models of planning and shows that many users do better with the latter.

DiamondHelp is a generic collaborative task guidance system motivated by the current usability crisis in high-tech home products. It combines an application-independent conversational interface (adapted from online chat programs) with an application-specific direct-manipulation interface. DiamondHelp is implemented in Java and uses Collagen for representing and using task models.

We describe an intelligent personal assistant that has been developed to aid a busy knowledge worker in managing time commitments and performing tasks. The design of the system was motivated by the complementary objectives of (1) relieving the user of routine tasks, thus allowing her to focus on tasks that critically require human problem-solving skills, and (2) intervening in situations where cognitive overload leads to oversights or mistakes by the user. The system draws on a diverse set of AI technologies that are linked within a Belief-Desire-Intention (BDI) agent system. Although the system provides a number of automated functions, the overall framework is highly user centric in its support for human needs, responsiveness to human inputs, and adaptivity to user working style and preferences.

Mixed-initiative systems are a popular approach to building intelligent systems that can collaborate naturally and effectively with people. But true collaborative behavior requires an agent to possess a number of capabilities, including reasoning, communication, planning, execution, and learning. We describe an integrated approach to the design and implementation of a collaborative problem-solving assistant based on a formal theory of joint activity and a declarative representation of tasks. This approach builds on prior work by us and by others on mixed-initiative dialogue and planning systems.

Conversational dialogue is an oft-cited example of mixed-initiative interaction, referring to the ability of each participant in a dialogue to take initiative to guide or add to a discussion. Endowing an automated dialogue system communicate, and coordinate with with the ability both to take initiative ("What In the course like to add a side trip.") However, of efforts to achieve goals while immersed mixed-initiative interaction extends beyond in shared context. We continue to engage spoken conversations to include a broad spectrum one another in efficient, tightly woven of collaborative problem solving marked collaborations, reasoning with remarkable efficiency by an interleaving of contributions by different about the beliefs, preferences, intentions, participants. Mastering mixed-initiative interaction poses The inferences underlying successful collaborations a constellation of fascinating challenges and typically stream in such an effortless opportunities for AI researchers.

Mixed-initiative assistants are agents that interact seamlessly with humans to extend their problem-solving capabilities or provide new capabilities. Developing such agents requires the synergistic integration of many areas of AI, including knowledge representation, problem solving and planning, knowledge acquisition and learning, multiagent systems, discourse theory, and human-computer interaction. This paper introduces seven aspects of mixed-initiative reasoning (task, control, awareness, communication, personalization, architecture, and evaluation) and discusses them in the context of several state-of-the-art mixed-initiative assistants. The goal is to provide a framework for understanding and comparing existing mixed-initiative assistants and for developing general design principles and methods.

This paper introduces and analyzes a battery of inference models for the problem of semantic role labeling: one based on constraint satisfaction, and several strategies that model the inference as a meta-learning problem using discriminative classifiers. These classifiers are developed with a rich set of novel features that encode proposition and sentence-level information. To our knowledge, this is the first work that: (a) performs a thorough analysis of learning-based inference models for semantic role labeling, and (b) compares several inference strategies in this context. We evaluate the proposed inference strategies in the framework of the CoNLL-2005 shared task using only automatically-generated syntactic information. The extensive experimental evaluation and analysis indicates that all the proposed inference strategies are successful -they all outperform the current best results reported in the CoNLL-2005 evaluation exercise- but each of the proposed approaches has its advantages and disadvantages. Several important traits of a state-of-the-art SRL combination strategy emerge from this analysis: (i) individual models should be combined at the granularity of candidate arguments rather than at the granularity of complete solutions; (ii) the best combination strategy uses an inference model based in learning; and (iii) the learning-based inference benefits from max-margin classifiers and global feedback.