Handshake request motion model with an approaching human for a handshake robot system

Jindai, Mitsuru; Ota, Shunsuke; Ikemoto, Yusuke; Sasaki, Tohru

doi:10.1109/iccis.2015.7274632

Cited by 10 publications

(23 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This could be explained by considering that the human was always in charge of the handshake initiation (via making contact with the robot hand), while the robot was static in the initial position P 0 , which would likely have contributed to the perception of the interaction leader. This perception could be changed, for example, by performing an approach motion of the arm before requesting the handshake from the subject, as in Jindai et al (2015). We also hypothesize that the term T ant plays a role on both the ''leadership'' and the ''responsiveness'' aspects of the handshake.…”

Section: Discussionmentioning

confidence: 98%

See 1 more Smart Citation

On the role of stiffness and synchronization in human–robot handshaking

Mura

Knoop

Catalano

et al. 2020

The International Journal of Robotics Research

View full text Add to dashboard Cite

This article presents a system for soft human–robot handshaking, using a soft robot hand in conjunction with a lightweight and impedance-controlled robot arm. Using this system, we study how different factors influence the perceived naturalness, and give the robot different personality traits. Capitalizing on recent findings regarding handshake grasp force regulation, and on studies of the impedance control of the human arm, we investigate the role of arm stiffness as well as the kinesthetic synchronization of human and robot arm motions during the handshake. The system is implemented using a lightweight anthropomorphic arm, with a Pisa/IIT Softhand wearing a sensorized silicone glove as the end-effector. The robotic arm is impedance-controlled, and its stiffness changes according to different laws under investigation. An internal observer is employed to synchronize the human and robot arm motions. Thus, we simulate both active and passive behavior of the robotic arm during the interaction. Using the system, studies are conducted where 20 participants are asked to interact with the robot, and then rate the perceived quality of the interaction using Likert scales. Our results show that the control of the robotic arm kinesthetic behavior does have an effect on the interaction with the robot, in term of its perceived personality traits, responsiveness, and human-likeness. Our results pave the way towards robotic systems that are capable of performing human–robot interactions in a more human-like manner, and with personality.

show abstract

Section: Discussionmentioning

confidence: 98%

“…Moreover, Avraham et al (2012) and Papageorgiou and Doulgeri (2015) presented examples of different motion models for consensus handshaking. Related to this line of research, Jindai et al (2015) focused on a robot simulating also the approaching motions before the handshake, that is, before the hands make contact.…”

Section: Introductionmentioning

confidence: 99%

On the role of stiffness and synchronization in human–robot handshaking

Mura

Knoop

Catalano

et al. 2020

The International Journal of Robotics Research

View full text Add to dashboard Cite

show abstract

“…This case is harder from a control perspective, because it requires coordination with the human hand, the robot needs to produce natural looking motions and it still involves physical contact as in the previous case. The control of robotic arm movement is investigated in [8], [9], but these papers do not control humanoid hands and do not observe complex contact dynamics. Furthermore, we investigate the possibility of performing different hand interactions, such as hand claps or finger touches, which are relatively under-researched.…”

Section: A Human Hand Interactionmentioning

confidence: 99%

“…We use the same hand model as stand-in for a human hand for convenience, since this model is easy to pose in different configurations. However, this model can be replaced with a human hand model, which should not influence the results of our experiments since the hand is not actuated, as usually assumed for a hand requesting an interaction [8], [9]. The robot models are taken from the OpenAI gym framework [13].…”

Section: B Simulation Environmentmentioning

confidence: 99%

“…In particular, we are interested in learning a control policy for diverse hand interactions, such as handshakes or hand claps. The handshake is the most common greeting gesture throughout the world, therefore it has received a lot of attention in the robotics community [5], [6], [7], [8], [9]. In this paper, we present a method for training a control policy for human-robot hand interactions, using data from human demonstrations in combination with deep reinforcement learning.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Demonstration-Guided Deep Reinforcement Learning of Control Policies for Dexterous Human-Robot Interaction

Christen

Stevšić

Hilliges

2019

2019 International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

In this paper, we propose a method for training control policies for human-robot interactions such as handshakes or hand claps via Deep Reinforcement Learning. The policy controls a humanoid Shadow Dexterous Hand, attached to a robot arm. We propose a parameterizable multi-objective reward function that allows learning of a variety of interactions without changing the reward structure. The parameters of the reward function are estimated directly from motion capture data of human-human interactions in order to produce policies that are perceived as being natural and human-like by observers. We evaluate our method on three significantly different hand interactions: handshake, hand clap and finger touch. We provide detailed analysis of the proposed reward function and the resulting policies and conduct a large-scale user study, indicating that our policy produces natural looking motions.

show abstract