Interaction primitives for human-robot cooperation tasks

By Heni Ben Amor, Gerhard Neumann, Sanket Kamthe, Oliver Kroemer, Jan Peters. 2014

233 citations recorded by the source corpus.

To engage in cooperative activities with human partners, robots have to possess basic interactive abilities and skills. However, programming such interactive skills is a challenging task, as each interaction partner can have different timing or an alternative way of executing movements. In this paper, we propose to learn interaction skills by observing how two humans engage in a similar task. To this end, we introduce a new representation called Interaction Primitives. Interaction primitives build on the framework of dynamic motor primitives (DMPs) by maintaining a distribution over the parameters of the DMP. With this distribution, we can learn the inherent correlations of cooperative activities which allow us to infer the behavior of the partner and to participate in the cooperation. We will provide algorithms for synchronizing and adapting the behavior of humans and robots during joint physical activities.

Subfields: Artificial Intelligence, Control and Systems Engineering, Social Psychology, Reinforcement Learning

How researchers see this paper

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

In Artificial Intelligence

1. Sergey Levine
2. Pieter Abbeel
3. Gerhard Neumann
4. Jan Peters
5. J. Bagnell
6. Chelsea Finn
7. Sonia Chernova
8. Harish Chaandar Ravichandar

In Control and Systems Engineering

1. Sergey Levine
2. Pieter Abbeel
3. Jan Peters
4. Gerhard Neumann
5. J. Bagnell
6. Harish Chaandar Ravichandar
7. Sonia Chernova
8. Aude Gemma Billard

In Social Psychology

1. Pieter Abbeel
2. J. Bagnell
3. Jan Peters
4. Sonia Chernova
5. Aude Gemma Billard
6. Gerhard Neumann
7. Sergey Levine
8. Stefan Schaal

In Reinforcement Learning

1. Sergey Levine
2. Pieter Abbeel
3. Jan Peters
4. Gerhard Neumann
5. Chelsea Finn
6. J. Bagnell
7. Shimon Whiteson
8. Shixiang Shane Gu

Citations in the corpus

• Learning Dynamic Robot-to-Human Object Handover from Human Feedback (2016, background): “To represent robot skills, policy search typically makes use of parametrized controllers, such as dynamic movement primitives [16] or interaction primitives [2].”
• An Algorithmic Perspective on Imitation Learning (2018, methodology): “For example, Amor et al. [2014] proposed to model the joint distribution of DMP parameters and generalize learned motion in human-robot interaction scenarios.”
• Recent Advances in Robot Learning from Demonstration (2020, background): “Indeed, several LfD approaches have been proposed to learn how to interact with a human user (62, 63, 69, 72, 75, 80, 105).”
• Kernelized movement primitives (2017, unknown): “For instance, a variant of policy improvement with path integrals (Buchli et al. 2011) was employed to optimize the movement pattern of DMP (Stulp and Sigaud 2013).”