2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2018
DOI: 10.1109/iros.2018.8594103
|View full text |Cite
|
Sign up to set email alerts
|

Probabilistic Learning of Torque Controllers from Kinematic and Force Constraints

Abstract: When learning skills from demonstrations, one is often required to think in advance about the appropriate task representation (usually in either operational or configuration space). We here propose a probabilistic approach for simultaneously learning and synthesizing torque control commands which take into account task space, joint space and force constraints. We treat the problem by considering different torque controllers acting on the robot, whose relevance is learned probabilistically from demonstrations. … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
40
0

Year Published

2018
2018
2023
2023

Publication Types

Select...
4
4

Relationship

4
4

Authors

Journals

citations
Cited by 25 publications
(40 citation statements)
references
References 23 publications
0
40
0
Order By: Relevance
“…Since our approach employs the kernel treatment instead of the explicit basis functions, it enables the convenient extension to the learning of complex and high-dimensional trajectories. In the future, we plan to apply our approach to the concurrent imitation learning in Cartesian space and joint space [33], [34].…”
Section: Discussionmentioning
confidence: 99%
“…Since our approach employs the kernel treatment instead of the explicit basis functions, it enables the convenient extension to the learning of complex and high-dimensional trajectories. In the future, we plan to apply our approach to the concurrent imitation learning in Cartesian space and joint space [33], [34].…”
Section: Discussionmentioning
confidence: 99%
“…where Γ p are weight matrices that regulate the contribution of each individual controller. Examples of Γ p found in the literature include scalar terms that maximize external rewards [20] and precision matrices, either computed from covariance [6], [19] or uncertainty [7]. Equation (12) has an analytical solution given byû =Σ u P p=1 Γ p u p , wherê…”
Section: Fusing Optimal Controllersmentioning
confidence: 99%
“…In recent work [6], [7], we discussed a fundamental difference between the type of variance encapsulated by the predictions of classical probabilistic techniques, particularly Gaussian mixture regression (GMR) and Gaussian process regression (GPR) [8]. We showed that the variance predicted by these two techniques has distinct, complementary interpretations.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…In [10], Umlauft et al discuss the differences between variance being interpreted as uncertainty and variability. The topic is also covered in [11], where the different notions are exploited in scenarios that require the combination of different controllers, and in [12], in the context of robot dynamics with multiple additive noise sources. The first contribution of the present work is the exploitation of the notion of variance as uncertainty to regulate impedance gains and render the robot compliant when uncertain about its actions.…”
Section: Related Workmentioning
confidence: 99%