Abstract
In many cooperative tasks between a human and a robotic assistant, the human guides the robot by exerting forces, either through direct physical interaction or indirectly via a jointly manipulated object. These physical forces perturb the robot's behavior execution and need to be compensated for in order to successfully complete such tasks. Typically, this problem is tackled by means of special purpose force sensors which are, however, not available on many robotic platforms. In contrast, we propose a machine learning approach based on sensor data, such as accelerometer and pressure sensor information. In the training phase, a statistical model of behavior execution is learned that combines Gaussian Process Regression with a novel periodic kernel. During behavior execution, predictions from the statistical model are continuously compared with stability parameters derived from current sensor readings. Differences between predicted and measured values exceeding the variance of the statistical model are interpreted as guidance information and used to adapt the robot's behavior. Several examples of cooperative tasks between a human and a humanoid NAO robot demonstrate the feasibility of our approach.
| Original language | English (US) |
|---|---|
| Title of host publication | IEEE-RAS International Conference on Humanoid Robots |
| Publisher | IEEE Computer Society |
| Pages | 124-129 |
| Number of pages | 6 |
| Volume | 2015-February |
| Edition | February |
| DOIs | |
| State | Published - Feb 3 2015 |
| Externally published | Yes |
| Event | 2013 13th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2013 - Atlanta, United States Duration: Oct 15 2013 → Oct 17 2013 |
Other
| Other | 2013 13th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2013 |
|---|---|
| Country | United States |
| City | Atlanta |
| Period | 10/15/13 → 10/17/13 |
Fingerprint
ASJC Scopus subject areas
- Artificial Intelligence
- Computer Vision and Pattern Recognition
- Hardware and Architecture
- Human-Computer Interaction
- Electrical and Electronic Engineering
Cite this
Inferring guidance information in cooperative human-robot tasks. / Berger, Erik; Vogt, David; Haji-Ghassemi, Nooshin; Jung, Bernhard; Ben Amor, Hani.
IEEE-RAS International Conference on Humanoid Robots. Vol. 2015-February February. ed. IEEE Computer Society, 2015. p. 124-129 7029966.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Inferring guidance information in cooperative human-robot tasks
AU - Berger, Erik
AU - Vogt, David
AU - Haji-Ghassemi, Nooshin
AU - Jung, Bernhard
AU - Ben Amor, Hani
PY - 2015/2/3
Y1 - 2015/2/3
N2 - In many cooperative tasks between a human and a robotic assistant, the human guides the robot by exerting forces, either through direct physical interaction or indirectly via a jointly manipulated object. These physical forces perturb the robot's behavior execution and need to be compensated for in order to successfully complete such tasks. Typically, this problem is tackled by means of special purpose force sensors which are, however, not available on many robotic platforms. In contrast, we propose a machine learning approach based on sensor data, such as accelerometer and pressure sensor information. In the training phase, a statistical model of behavior execution is learned that combines Gaussian Process Regression with a novel periodic kernel. During behavior execution, predictions from the statistical model are continuously compared with stability parameters derived from current sensor readings. Differences between predicted and measured values exceeding the variance of the statistical model are interpreted as guidance information and used to adapt the robot's behavior. Several examples of cooperative tasks between a human and a humanoid NAO robot demonstrate the feasibility of our approach.
AB - In many cooperative tasks between a human and a robotic assistant, the human guides the robot by exerting forces, either through direct physical interaction or indirectly via a jointly manipulated object. These physical forces perturb the robot's behavior execution and need to be compensated for in order to successfully complete such tasks. Typically, this problem is tackled by means of special purpose force sensors which are, however, not available on many robotic platforms. In contrast, we propose a machine learning approach based on sensor data, such as accelerometer and pressure sensor information. In the training phase, a statistical model of behavior execution is learned that combines Gaussian Process Regression with a novel periodic kernel. During behavior execution, predictions from the statistical model are continuously compared with stability parameters derived from current sensor readings. Differences between predicted and measured values exceeding the variance of the statistical model are interpreted as guidance information and used to adapt the robot's behavior. Several examples of cooperative tasks between a human and a humanoid NAO robot demonstrate the feasibility of our approach.
UR - http://www.scopus.com/inward/record.url?scp=84937855652&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84937855652&partnerID=8YFLogxK
U2 - 10.1109/HUMANOIDS.2013.7029966
DO - 10.1109/HUMANOIDS.2013.7029966
M3 - Conference contribution
AN - SCOPUS:84937855652
VL - 2015-February
SP - 124
EP - 129
BT - IEEE-RAS International Conference on Humanoid Robots
PB - IEEE Computer Society
ER -