Teaching and learning of robot tasks via observation of human performance

By Rüdiger Dillmann. Robotics Auton. Syst., 2004

345 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/4ca073b43c04c5215d131ac829adc57e81476d8e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Control and Systems Engineering

Citations in the corpus

Recent Advances in Robot Learning from Demonstration (2020, background): “The third demonstration approach is for the robot to learn from passive observations of the user (36, 37, 38).”
Imitative Reinforcement Learning for Soccer Playing Robots (2006, background): “Imitation learning is a well established concept in robotics [4,5,6,7,8,9].”
A syntactic approach to robot imitation learning using probabilistic activity grammars (2013, neutral): “Our approach shares the concept of imitation learning presented in the Handbook of Robotics (Chapter 59) [1], as well as in [2–5] where a robot learns a new task directly from human demonstration without the need of extensive reprogramming.”
Task-Specific Generalization of Discrete and Periodic Dynamic Movement Primitives (2010, background): “One of the most successful paradigms that can be used for this purpose is imitation or robot programming by demonstration [1], [2].”
One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learning (2018, background): “R ELATED W ORK Most imitation learning and learning from demonstration methods operate at the level of conﬁguration-space trajectories [44, 2], which are typically collected using kinesthetic teaching [32, 1], teleoperation [8, 35, 62], or sensors on the demonstrator [11, 9, 7, 21].”
Evaluation of robot imitation attempts: comparison of the system's and the human's perspectives (2006, neutral): “The major directions of work have been using such behaviours with humans for adaptive learning [8, 12, 6, 3, 7] and also allowing the robots to be able to engage in richer, more natural social interactions [5, 11, 4].”
Dynamical System Modulation for Robot Learning via Kinesthetic Demonstrations (2008, methodology): “This has been done using HMMs [15] or knowledgebased systems [16].”
ARMAR-III: An Integrated Humanoid Platform for Sensory-Motor Control (2006, background): “In particular, we address the integration of motor, perception and cognition components such as multimodal human-humanoid interaction and humanhumanoid cooperation in order to be able to demonstrate robot tasks in a kitchen environment as a prototypical humancentered one [1].”
Bootstrapping humanoid robot skills by extracting semantic representations of human-like activities from virtual reality (2014, background): “Regarding the problem of human activity recognition several challenges have to be addressed, for instance: automatic segmentation of human motions [9], [10], identification of important features of the motion [11], definition of the importance of the object(s) to the task [12], as well as the defini”
Incremental learning of gestures by imitation in a humanoid robot (2007, background): “Robot learning by imitation, also referred to as robot programming by demonstration, explores novel means of implicitly teaching a robot new motor skills [5, 10, 20].”
Kinesthetic teaching via Fast Marching Square (2012, methodology): “In [4] programming by demonstration is used by teaching a robot to carry out specific tasks such as loading the dishwasher or taking out bottles from the refrigerator.”
A connectionist architecture for view-independent grip-aperture computation. (2008, background): “…investigations notably concern mechanisms that enable robotic agents to acquire new behaviours by observing and generalizing the behaviour of other agents ([Dillmann, 2005] and [Kruger, 2007]), and feed-back mechanisms that enable agents to control their own behaviour (Desmurget and Grafton, 2000).”
Human–robot interaction review and challenges on task planning and programming (2016, methodology): “The use of a vision system and data gloves (Dillmann 2004), as well as active tracking mechanisms that recognised ﬁ nger motions, were implemented in Perrin, Cassinelli, and Ishikawa (2004).”
Incremental Learning, Clustering and Hierarchy Formation of Whole Body Motion Patterns using Adaptive Hidden Markov Chains (2008, background): “A common paradigm is Programming by Demonstration (PbD) (Dillmann et al. 1999 Dillmann 2004 Bernardin et al. 2005).”
On Human Action (2011, background): “In robotics, generative implies an explicit modeling of each of the categories of action: the notion of generative models often embraces representations that not only support the recognition of an action but also permit the robot to synthesize such an action itself [6, 11, 43, 44].”