For tasks such as dressing assistance, robots should be able to adapt to different user morphologies, preferences and requirements. We propose a programming by demonstration method to efficiently learn and adapt such skills. Our method encodes sensory information (relative to the human user) and motor commands (relative to the robot actuation) as a joint distribution in a hidden semi-Markov mod...

Programming by Demonstration, or PBD, is an exciting and developing branch of HCI research. With PBD techniques, end-users can add functionality to their environments without programming in the conventional sense. Virtually all research into PBD, however, presumes that the event history is a linear sequence of user actions. This paper challenges that notion by introducing Hierarchical Event His...

Programming-by-demonstration (PBD) systems learn tasks by watching the user perform them. CIMA1 is an interactive learning system for modeling the data selected and modified by a user as he or she undertakes a task. Part of a PBD system, CIMA is invoked when user actions are matched to find a common description of their “operands.” Although the system’s interfaces to users and applications are ...

CHI ’15 Workshop on Mobile Collocated Interactions Abstract Mobile devices are generally mostly used for relatively simple tasks rather than productivity tasks such as word processing and creating presentations and spreadsheets common to desktop environments. Despite increasingly more powerful hand-held and wearable devices, but due to small screen real estate and limitations of touch interacti...

A decent linear space (DLS) is a linear space (or PBD of index 1) without lines of size 1 or 2; see Beth, Jungnickel, and Lenz [l] for background and definitions. We denote the maximal line size of a DLS by k and write DLS(k), then; if we also want to specify the number u of points, we use the notation DLS(k; v). We always assume all linear spaces to be non-trivial, i.e., u # k. We shall be con...

or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works, must be obtained from the IEEE. Contact: IEEE Intellectual Property Rights Office / IEEE Service Center / 445 Hoes Lane / Piscataway, NJ 08855-1331 / phone: (732) 562-3966 / fax: (732) 981-8062 In Proc. IEEE ICRA, Workshop on The Social Mechan...

This paper presents a pattern recognition approach to identifying contact formations from force sensor signals. The approach is sensor-based and does not use geometric models of the workpieces. The design of a fuzzy classiier is described, where membership functions are generated automatically from training data. The technique is demonstrated using supervised learning. Test results are included...

Two algorithms that combine Brownian dynamics (BD) simulations with mean-field partial differential equations (PDEs) are presented. This PDE-assisted Brownian dynamics (PBD) methodology provides exact particle tracking data in parts of the domain, whilst making use of a mean-field reaction-diffusion PDE description elsewhere. The first PBD algorithm couples BD simulations with PDEs by randomly ...

Programming by demonstratioti‘(PBD) is a term that describes a wide range of end-user programming techniques that generate code from the user’s examples. Most systems that use PBD rely on knowledge of a specijic application (domain knowledge) and on ad hoc inferencing techniques to make sense of the user’s actions. This paper argues that it is possible and practical to build a PED system that i...

Scene-level Programming by Demonstration (PbD) is faced with an important challenge perceptual uncertainty. Addressing this problem, we present a scenelevel PbD paradigm that programs robots to perform goal-directed manipulation in unstructured environments with grounded perception. Scene estimation is enabled by our discriminatively-informed generative scene estimation method (DIGEST). Given s...