Abstract
Reliable physical interaction is essential for many important challenges in robotic manipulation. In this paper, we consider Constrained Object Manipulations tasks (COM), i.e. tasks for which constraints are imposed on the grasped object rather than on the robot’s configuration. To enable robust physical interaction with the environment, this paper presents a manifold learning approach to encode the COM task as a vector field. This representation enables an intuitive task-consistent adaptation based on an object-level impedance controller. Simulations and experimental evaluations demonstrate the effectiveness of our approach for several typical COM tasks, including dexterous manipulation and contour following.
| Original language | English |
|---|---|
| Pages (from-to) | 159-174 |
| Number of pages | 16 |
| Journal | Autonomous Robots |
| Volume | 42 |
| Issue number | 1 |
| DOIs | |
| Publication status | Published - Jan 1 2018 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Artificial Intelligence
Cite this
Learning task manifolds for constrained object manipulation. / Li, Miao; Tahara, Kenji; Billard, Aude.
In: Autonomous Robots, Vol. 42, No. 1, 01.01.2018, p. 159-174.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Learning task manifolds for constrained object manipulation
AU - Li, Miao
AU - Tahara, Kenji
AU - Billard, Aude
PY - 2018/1/1
Y1 - 2018/1/1
N2 - Reliable physical interaction is essential for many important challenges in robotic manipulation. In this paper, we consider Constrained Object Manipulations tasks (COM), i.e. tasks for which constraints are imposed on the grasped object rather than on the robot’s configuration. To enable robust physical interaction with the environment, this paper presents a manifold learning approach to encode the COM task as a vector field. This representation enables an intuitive task-consistent adaptation based on an object-level impedance controller. Simulations and experimental evaluations demonstrate the effectiveness of our approach for several typical COM tasks, including dexterous manipulation and contour following.
AB - Reliable physical interaction is essential for many important challenges in robotic manipulation. In this paper, we consider Constrained Object Manipulations tasks (COM), i.e. tasks for which constraints are imposed on the grasped object rather than on the robot’s configuration. To enable robust physical interaction with the environment, this paper presents a manifold learning approach to encode the COM task as a vector field. This representation enables an intuitive task-consistent adaptation based on an object-level impedance controller. Simulations and experimental evaluations demonstrate the effectiveness of our approach for several typical COM tasks, including dexterous manipulation and contour following.
UR - http://www.scopus.com/inward/record.url?scp=85021051527&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85021051527&partnerID=8YFLogxK
U2 - 10.1007/s10514-017-9643-z
DO - 10.1007/s10514-017-9643-z
M3 - Article
AN - SCOPUS:85021051527
VL - 42
SP - 159
EP - 174
JO - Autonomous Robots
JF - Autonomous Robots
SN - 0929-5593
IS - 1
ER -