Supervised Learning of Gesture-Action Associations for Human-Robot Collaboration

Shukla, Dadhichi; Erkent, Özgür; Piater, Justus

doi:10.1109/fg.2017.97

Cited by 4 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unlike previous works (Lenz et al, 2008 ; Myagmarjav and Sridharan, 2015 ), in PIL the mapping between an instruction and a robot action materializes during the task, therefore bypassing a training phase. We demonstrated in our previous work (Shukla et al, 2017b ) the advantage of learning semantics of instructions while performing the task using PIL over pre-trained associations.…”

Section: Discussionmentioning

confidence: 86%

“…Rather, the gesture-action associations are learnt incrementally on the fly, an advantage in terms of total interaction time. In our previous work (Shukla et al, 2017b ) we showed benefits of proactive learning over a pre-trained system in overall interaction time.…”

Section: Introductionmentioning

confidence: 73%

“…Additionally, we also present results of both IRL methods at a higher learning rate α = 0.9. In our previous work (Shukla et al, 2017b ) we showed how the robot's gaze facilitates communication and speeds up completion of the task. Therefore, for a fair comparison we incorporate gaze into both IRL methods.…”

Section: Methodsmentioning

confidence: 93%

See 2 more Smart Citations

Learning Semantics of Gestural Instructions for Human-Robot Collaboration

2018

Self Cite

View full text Add to dashboard Cite

Designed to work safely alongside humans, collaborative robots need to be capable partners in human-robot teams. Besides having key capabilities like detecting gestures, recognizing objects, grasping them, and handing them over, these robots need to seamlessly adapt their behavior for efficient human-robot collaboration. In this context we present the fast, supervised Proactive Incremental Learning (PIL) framework for learning associations between human hand gestures and the intended robotic manipulation actions. With the proactive aspect, the robot is competent to predict the human's intent and perform an action without waiting for an instruction. The incremental aspect enables the robot to learn associations on the fly while performing a task. It is a probabilistic, statistically-driven approach. As a proof of concept, we focus on a table assembly task where the robot assists its human partner. We investigate how the accuracy of gesture detection affects the number of interactions required to complete the task. We also conducted a human-robot interaction study with non-roboticist users comparing a proactive with a reactive robot that waits for instructions.

show abstract

Section: Discussionmentioning

confidence: 86%

Section: Introductionmentioning

confidence: 73%