Vincent Aleven Intelligent tutoring systems (ITS) are highly effective in supporting student learning, but are difficult to build. The Cognitive Tutor Authoring Tools (CTAT) project started over 6 years ago with the goals of making it easier for experienced programmers, and possible for non-programmers to create an ITS. CTAT supports tutor building through programming by demonstration, an appro...

We are proposing a supervised learning approach in robot force control which enables robot programming by demonstration through an operator. The learning element of the controller is based on a neural network prestructured to represent knowledge in terms of rules. Tasks are demonstrated in a virtual environment. Visual feedback using sensor ball devices and graphic display of forces as well as ...

Robust methods for representing, generalizing, and sharing knowledge across various robotics systems and configurations are important in many domains of robotics research and application. In this paper we present a method for modeling tasks and robot skills to simplify the programming and reuse of knowledge between robots in manufacturing environments. Specifically, we propose an assembly taxon...

We present three descriptors of texture feature of a region. Namely, the homogeneous texture descriptor (HTD), the edge histogram descriptor (EHD), and the perceptual browsing descriptor (PBD). They are currently included in the Committee Draft of the MPEG-7 Visual (ISO/IEC 15938-3). Each descriptor has a unique functionality and application domain. HTD and EHD describe statistical distribution...

This paper deals with programming-bydemonstration of robot manipulators, where a robot program is created automatically from human demonstrations despite morphological differences in the hand/arm structure. To find a correspondence between observed human motions and desired robot hand motions, we represent the task in hand state space. The hand state definition is perception-invariant and valid...

We trace the tragicomic courtship between programming by demonstration and machine learning, two fields that seem made for each other but have never quite got together. A long-term historical perspective illuminates the twin roles of interaction and sequence learning in programming by demonstration, and reveals the parallel growth of machine learning as a research area in its own right. A view ...

Traditional machine learning algorithms have failed to serve the needs of systems for Programming by Demonstration (PBD), which require interaction with a user (a teacher) and a task environment. We argue that traditional learning algorithms fail for two reasons: they do not cope with the ambiguous instructions that users provide in addition to examples; and their learning criterion requires on...

Programming by Demonstration (PbD) is an intuitive method to program a robot. The user, acting as a teacher or programmer, shows how a particular task can be performed. The demonstration is monitored using an interface device that allows the measurement and recording of both the applied commands as well as the data simultaneously perceived by robot's sensors. This paper considers the problems o...

We present an approach to learning procedural knowledge by demonstration called similarity-based alignment and generalization. Key to our approach is the ability to induce complex procedure structure (loops and conditional branches) by aligning multiple unannotated demonstrations of a procedure. We present an implemented instance of a similarity-based alignment and generalization algorithm that...

Currently, most demonstrationai systems are built from scratch, typically by individuals, and are all quite unique and distinct from each other. It is still desirable to build more such small-scale systems in the future. However, the key to a larger-scale, collaborative and interdisciplinary effort is in defining a common representation for user tasks, and in defining a common i put format for ...