https://web.cs.wpi.edu/~rail/files/ArgallEtAlRAS09.pdf

A survey of robot learning from demonstration

We present a programming-by-demonstration framework for generically extracting the relevant features of a given task and for addressing the problem of generalizing the acquired knowledge to different contexts. We validate the architecture through a series of experiments, in which a human demonstrator teaches a humanoid robot simple manipulatory tasks. A probability-based estimation of the relevance is suggested by first projecting the motion data onto a generic latent space using principal component analysis. The resulting signals are encoded using a mixture of Gaussian/Bernoulli distributions (Gaussian mixture model/Bernoulli mixture model). This provides a measure of the spatio-temporal correlations across the different modalities collected from the robot, which can be used to determine a metric of the imitation performance. The trajectories are then generalized using Gaussian mixture regression. Finally, we analytically compute the trajectory which optimizes the imitation metric and use this to generalize the skill to different contexts

/pdf/on-learning-representing-and-generalizing-a-task-in-a-1ynfj1gzbh.pdf

On Learning, Representing, and Generalizing a Task in a Humanoid Robot

Also referred to as learning by imitation, tutelage, or apprenticeship learning, Programming by Demonstration (PbD) develops methods by which new skills can be transmitted to a robot. This book examines methods by which robots learn new skills through human guidance. Taking a practical perspective, it covers a broad range of applications, including service robots. The text addresses the challenges involved in investigating methods by which PbD is used to provide robots with a generic and adaptive model of control. Drawing on findings from robot control, human-robot interaction, applied machine learning, artificial intelligence, and developmental and cognitive psychology, the book contains a large set of didactic and illustrative examples. Practical and comprehensive machine learning source codes are available on the books companion website: http://www.programming-by-demonstration.org

Robot Programming by Demonstration

This study reviews recently published scientific literature on the use of robotics in schools, in order to: (a) identify the potential contribution of the incorporation of robotics as educational tool in schools, (b) present a synthesis of the available empirical evidence on the educational effectiveness of robotics as an educational tool in schools, and (c) define future research perspectives concerning educational robotics. After systematically searching online bibliographic databases, ten relevant articles were located and included in the study. For each article, we analyze the purpose of the study, the content to be taught with the aid of robotics, the type of robot used, the research method used, and the sample characteristics (sample size, age range of students and/or level of education) and the results observed. The articles reviewed suggest that educational robotics usually acts as an element that enhances learning, however, this is not always the case, as there are studies that have reported situations in which there was no improvement in learning. The outcomes of the literature review are discussed in terms of their implications for future research, and can provide useful guidance for educators, practitioners and researchers in the area. Highlights? We performed a systematic review of studies with quantitative evidence of the use of robotics in schools. ? Studies indicate positive outcomes for teaching concepts related to the STEM areas. ? Nine important factors to increase the success of robotics as a teaching tool are presented. ? More research is needed about how to use robotics to develop new skills in students.

Exploring the educational potential of robotics in schools

This paper presents a method to learn discrete robot motions from a set of demonstrations. We model a motion as a nonlinear autonomous (i.e., time-invariant) dynamical system (DS) and define sufficient conditions to ensure global asymptotic stability at the target. We propose a learning method, which is called Stable Estimator of Dynamical Systems (SEDS), to learn the parameters of the DS to ensure that all motions closely follow the demonstrations while ultimately reaching and stopping at the target. Time-invariance and global asymptotic stability at the target ensures that the system can respond immediately and appropriately to perturbations that are encountered during the motion. The method is evaluated through a set of robot experiments and on a library of human handwriting motions.

/pdf/learning-stable-nonlinear-dynamical-systems-with-gaussian-23fwsbi8rv.pdf

Learning Stable Nonlinear Dynamical Systems With Gaussian Mixture Models

We present a system for robust robot skill acquisition from kinesthetic demonstrations. This system allows a robot to learn a simple goal-directed gesture and correctly reproduce it despite changes in the initial conditions and perturbations in the environment. It combines a dynamical system control approach with tools of statistical learning theory and provides a solution to the inverse kinematics problem when dealing with a redundant manipulator. The system is validated on two experiments involving a humanoid robot: putting an object into a box and reaching for and grasping an object.

/pdf/dynamical-system-modulation-for-robot-learning-via-1m00bg4knu.pdf

Dynamical System Modulation for Robot Learning via Kinesthetic Demonstrations

This paper presents an architecture for solving generically the problem of extracting the constraints of a given task in a programming by demonstration framework and the problem of generalizing the acquired knowledge to various contexts. We validate the architecture in a series of experiments, where a human demonstrator teaches a humanoid robot simple manipulatory tasks. First, the combined joint angles and hand path motions are projected into a generic latent space, composed of a mixture of Gaussians (GMM) spreading across the spatial dimensions of the motion. Second, the temporal variation of the latent representation of the motion is encoded in a hidden Markov model (HMM). This two-step probabilistic encoding provides a measure of the spatio-temporal correlations across the different modalities collected by the robot, which determines a metric of imitation performance. A generalization of the demonstrated trajectories is then performed using Gaussian mixture regression (GMR). Finally, to generalize skills across contexts, we compute formally the trajectory that optimizes the metric, given the new context and the robot's specific body constraints

/pdf/on-learning-the-statistical-representation-of-a-task-and-27dhov4fo4.pdf

On learning the statistical representation of a task and generalizing it to various contexts

This paper presents a prototype of 3 degrees-of-freedom articulated spine for the doll-shaped humanoid robot Robota. This work follows an approach that emphasizes the need for a high human-likeliness in both the external features of the robot and in the kinematics of its motions to enhance human-robot interactions. The design of a spinal cord for our humanoid robot satisfies both criteria in providing offering a smooth human-like parallel means of bending forward and sideways

Design of a Biomimetic Spine for the Humanoid Robot Robota

The goal of developing algorithms for programming robots by demonstration is to create an easy way of programming robots that can be accomplished by everyone. When a demonstrator teaches a task to a robot, he/she shows some ways of fulfilling the task, but not all the possibilities. The robot must then be able to reproduce the task even when unexpected perturbations occur. In this case, it has to learn a new solution. In this paper, we describe a system that allows a robot to re-learn constrained reaching tasks by combining the knowledge acquired during the demonstration, with that acquired though reinforcement learning.

F. Guenter

Papers

On Learning, Representing, and Generalizing a Task in a Humanoid Robot

Dynamical System Modulation for Robot Learning via Kinesthetic Demonstrations

On learning the statistical representation of a task and generalizing it to various contexts

Design of a Biomimetic Spine for the Humanoid Robot Robota

Using reinforcement learning to adapt an imitation task