Inferring subjective preferences on robot trajectories using EEG signals

Iwane, Fumiaki; Halvagal, Manu Srinath; Iturrate, Iñaki; Batzianoulis, Iason; Chavarriaga, Ricardo; Billard, Aude; Millán, José del R.

doi:10.1109/ner.2019.8717025

Cited by 18 publications

(10 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Users [0, 1,4,5,11,16] had the same event sequence: short and long board connections (shown in grey), connector and board connections (green), and shelf and board connections (yellow). Similarly, other groups of users like [12,13,15], and [3,9,10,14,17] also had the same event sequences. The assembly task is fairly complex; the 32 primary actions can be ordered in more than 24!…”

Section: B Analysis Of User Preferencesmentioning

confidence: 80%

“…Preferences can then be modelled as a feed-forward neural network that maps task metrics to the survey responses of users [14]. Human preferences for robot actions can also be measured from EEG signals [15], or from a short window of human arm motion [16]- [18] or gaze pattern [19]. Past work has also focused on incorporating user preferences in task assignment and scheduling, where the user preferences are included as a constraint [20], [21] or in the objective function [22] of the scheduling problem.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Two-Stage Clustering of Human Preferences for Action Prediction in Assembly Tasks

Nemlekar¹,

Modi²,

Gupta³

et al. 2021

Preprint

View full text Add to dashboard Cite

To effectively assist human workers in assembly tasks a robot must proactively offer support by inferring their preferences in sequencing the task actions. Previous work has focused on learning the dominant preferences of human workers for simple tasks largely based on their intended goal. However, people may have preferences at different resolutions: they may share the same high-level preference for the order of the sub-tasks but differ in the sequence of individual actions. We propose a two-stage approach for learning and inferring the preferences of human operators based on the sequence of sub-tasks and actions. We conduct an IKEA assembly study and demonstrate how our approach is able to learn the dominant preferences in a complex task. We show that our approach improves the prediction of human actions through cross-validation. Lastly we show that our two-stage approach improves the efficiency of task execution in an online experiment and demonstrate its applicability in a real-world robot-assisted IKEA assembly.

show abstract

Section: B Analysis Of User Preferencesmentioning

confidence: 80%

Section: Related Workmentioning

confidence: 99%

Two-Stage Clustering of Human Preferences for Action Prediction in Assembly Tasks

Nemlekar¹,

Modi²,

Gupta³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Chavarriaga et al (2007); Iturrate et al (2010b) and Yousefi et al (2018) have shown that ErrPs are elicited when the subject has to engage in tasks that require a high level of concentration (e.g., motor imagery or cognitive task). The presence of an ErrP when the subjects evaluate the actions of a robotic arm based only on individual subjective criteria was addressed by Iwane et al (2019). The possibility to detect ErrP in a virtual reality environment has been evaluated 2018) have additionally proposed automatically replacing characters with the second-best letter.…”

Section: Error-related Potentials In Brain-machine Interfacesmentioning

confidence: 99%

Error-Related Potentials in Reinforcement Learning-Based Brain-Machine Interfaces

Fidêncio

Klaes

Iossifidis

2022

Front. Hum. Neurosci.

View full text Add to dashboard Cite

The human brain has been an object of extensive investigation in different fields. While several studies have focused on understanding the neural correlates of error processing, advances in brain-machine interface systems using non-invasive techniques further enabled the use of the measured signals in different applications. The possibility of detecting these error-related potentials (ErrPs) under different experimental setups on a single-trial basis has further increased interest in their integration in closed-loop settings to improve system performance, for example, by performing error correction. Fewer works have, however, aimed at reducing future mistakes or learning. We present a review focused on the current literature using non-invasive systems that have combined the ErrPs information specifically in a reinforcement learning framework to go beyond error correction and have used these signals for learning.

show abstract

“…A smaller number of publications utilized informative EEG signals from users who were passively observing a moving robot. Examples include the identification of erroneous actions (Salazar-Gomez et al, 2017;Behncke et al, 2018) or user preferences for robot motion (Iwane et al, 2019;Kolkhorst et al, 2019a). While this allows to infer user judgment of robotic actions, it requires the robot to first perform a candidate action (e.g., moving to the object with the highest prior probability).…”

Section: Related Workmentioning

confidence: 99%

A Robust Screen-Free Brain-Computer Interface for Robotic Object Selection

et al. 2020

View full text Add to dashboard Cite

Brain signals represent a communication modality that can allow users of assistive robots to specify high-level goals, such as the object to fetch and deliver. In this paper, we consider a screen-free Brain-Computer Interface (BCI), where the robot highlights candidate objects in the environment using a laser pointer, and the user goal is decoded from the evoked responses in the electroencephalogram (EEG). Having the robot present stimuli in the environment allows for more direct commands than traditional BCIs that require the use of graphical user interfaces. Yet bypassing a screen entails less control over stimulus appearances. In realistic environments, this leads to heterogeneous brain responses for dissimilar objects-posing a challenge for reliable EEG classification. We model object instances as subclasses to train specialized classifiers in the Riemannian tangent space, each of which is regularized by incorporating data from other objects. In multiple experiments with a total of 19 healthy participants, we show that our approach not only increases classification performance but is also robust to both heterogeneous and homogeneous objects. While especially useful in the case of a screen-free BCI, our approach can naturally be applied to other experimental paradigms with potential subclass structure.

show abstract

Inferring subjective preferences on robot trajectories using EEG signals

Cited by 18 publications

References 8 publications

Two-Stage Clustering of Human Preferences for Action Prediction in Assembly Tasks

Two-Stage Clustering of Human Preferences for Action Prediction in Assembly Tasks

Error-Related Potentials in Reinforcement Learning-Based Brain-Machine Interfaces

A Robust Screen-Free Brain-Computer Interface for Robotic Object Selection

Contact Info

Product

Resources

About