2011
DOI: 10.1007/s10514-011-9234-3
|View full text |Cite
|
Sign up to set email alerts
|

Incremental kinesthetic teaching of motion primitives using the motion refinement tube

Abstract: We present an approach for kinesthetic teaching of motion primitives for a humanoid robot. The proposed teaching method starts with observational learning and applies iterative kinesthetic motion refinement using a forgetting factor. Kinesthetic teaching is supported by introducing the motion refinement tube, which represents an area of allowed motion refinement around the nominal trajectory. On the realtime control level, the kinesthetic teaching is handled by a customized impedance controller, which combines… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
124
0
6

Year Published

2011
2011
2024
2024

Publication Types

Select...
4
3
3

Relationship

3
7

Authors

Journals

citations
Cited by 159 publications
(130 citation statements)
references
References 44 publications
0
124
0
6
Order By: Relevance
“…1) Data collection: Kinesthetic teaching is often used for data collection in PbD [6], [13]. In kinesthetic teaching, a teacher physically holds the robot's end-effector for generating the required motion.…”
Section: A Deep-dmpmentioning
confidence: 99%
“…1) Data collection: Kinesthetic teaching is often used for data collection in PbD [6], [13]. In kinesthetic teaching, a teacher physically holds the robot's end-effector for generating the required motion.…”
Section: A Deep-dmpmentioning
confidence: 99%
“…Lee and Ott proposed an incremental learning approach for iterative motion refinement. Their approach combines kinesthetic teaching with impedance control and represents the skill using a Hidden Markov Model (HMM) [18]. Our proposed approach goes beyond prior work by enabling the user to refine the skill interactively both during and after the learning process.…”
Section: Related Workmentioning
confidence: 99%
“…(16) is computed online from the model parameters. The retrieved signal encapsulates variation and correlation information in the form of a probabilistic flow tube, see e.g., (Lee and Ott, 2011). Initialization: (µ k , Σ k , π k ) ← GMM encoding the initial configuration by Alg.2 repeat θ ← task space motion command by the surgeon J ← jacobian corresponding to current configuration q P i ← position of selected points to be controlled J i ← jacobian of the i-th point to be controlled …”
Section: M-stepmentioning
confidence: 99%