Adaptive Robot Assisted Therapy Using Interactive Reinforcement Learning

Tsiakas, Konstantinos; Dagioglou, Maria; Karkaletsis, Vangelis; Makedon, Fillia

doi:10.1007/978-3-319-47437-3_2

Cited by 23 publications

(46 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By the end of the assessment mode, the system has an indicative user model U M for the current user. The system can use this model to classify the user into one of the existing user models, loading the corresponding USP, following the assumption that similar user models result in similar user-specific policies [25]. This policy is loaded as the personalized training policy.…”

Section: Discussion and Future Workmentioning

confidence: 99%

“…Another limitation in designing RL agents for interactive systems, including the definition of a proper state-action space, is defining an appropriate reward function that serves the purpose of the system [23,24]. Our research is motivated by these challenges that arise when different types of users and feedback types are considered for real-time personalization using reinforcement learning [25]. To this end, we illustrate the proposed interactive learning and adaptation framework with a cognitive training task, investigating how interactive RL methods (learning from feedback) can be used to integrate human-generated feedback through EEG data (task engagement) and facilitate personalization.…”

Section: Reinforcement Learning For Socially-assistive Roboticsmentioning

confidence: 99%

See 1 more Smart Citation

Task Engagement as Personalization Feedback for Socially-Assistive Robots and Cognitive Training

2018

Self Cite

View full text Add to dashboard Cite

Socially-Assistive Robotics (SAR) has been extensively used for a variety of applications, including educational assistants, exercise coaches and training task instructors. The main goal of such systems is to provide a personalized and tailored session that matches user abilities and needs. While objective measures (e.g., task performance) can be used to adjust task parameters (e.g., task difficulty), towards personalization, it is essential that such systems also monitor task engagement to personalize their training strategies and maximize the effects of the training session. We propose an Interactive Reinforcement Learning (IRL) framework that combines explicit feedback (task performance) with implicit human-generated feedback (task engagement) to achieve efficient personalization. We illustrate the framework with a cognitive training task, describing our data-driven methodology (data collection and analysis, user simulation) towards designing our proposed real-time system. Our data analysis and the reinforcement learning experiments on real user data indicate that the integration of task engagement as human-generated feedback in the RL mechanism can facilitate robot personalization, towards a real-time personalized robot-assisted training system.

show abstract

Section: Discussion and Future Workmentioning

confidence: 99%

Section: Reinforcement Learning For Socially-assistive Roboticsmentioning

confidence: 99%

Task Engagement as Personalization Feedback for Socially-Assistive Robots and Cognitive Training

2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our previous work includes the development of multimodal interactive systems for adaptive cognitive and physical rehabilitation [18,30,29]. In this project, we propose to explore Interactive Reinforcement Learning (IRL) techniques using our proposed adaptation framework [29], to integrate user feedback (implicit or explicit) and human expertise feedback (e.g. coming from the HCPS human supervisor) [28].…”

Section: Interactive Robot Learningmentioning

confidence: 99%

An Interactive Multisensing Framework for Personalized Human Robot Collaboration and Assistive Training Using Reinforcement Learning

Tsiakas

Papakostas

Theofanidis

et al. 2017

Proceedings of the 10th International Conference on PErvasive Technologies Related to Assistive Environments

Self Cite

View full text Add to dashboard Cite

There is a recent trend of research and applications of Cyber-Physical Systems (CPS) in manufacturing to enhance humanrobot collaboration and production. In this paper, we propose a CPS framework for personalized Human-Robot Collaboration and Training to promote safe human-robot collaboration in manufacturing environments. We propose a human-centric CPS approach that focuses on multimodal human behavior monitoring and assessment, to promote human worker safety and enable human training in Human-Robot Collaboration tasks. We present the architecture of our proposed system, our experimental testbed and our proposed methods for multimodal physiological sensing, human state monitoring and interactive robot adaptation, to enable personalized interaction. CCS Concepts •Human-centered computing ! Human computer interaction (HCI); •Computing methodologies ! Artificial intelligence; •Computer systems organization ! Embedded and cyber-physical systems;

show abstract

“…In HMI trials, the relatively small amount of recorded data compared to other implementations of machine learning means that a direct RL approach is generally favored, instead of a data-driven deep learning-based approach. Examples of this approach can be seen in the game-like robotic rehabilitation HMIs implemented in [3], [15], in which RL allows for in-game adaptations of the physical system's response so the user remains motivated and engaged.…”

Section: Introductionmentioning

confidence: 99%

A Reinforcement Learning Assisted Eye-Driven Computer Game Employing a Decision Tree-Based Approach and CNN Classification

et al. 2021

View full text Add to dashboard Cite

Human-Machine Interfaces employing biosignal-based inputs are hard to translate to reallife applications, in part because of the difficulty of developing generalized models to classify physiological events representing a user's actions. In the proposed framework, an Electrooculography (EOG)-based game is operated through a pipeline of decision methods. These include a user-independent classification model of eye movements using a Convolutional Neural Network (CNN), which is fed with images created from signal windows, and an Ensemble of Utility Decision Networks (EUDN), which moderates the impact of oftentimes conflicting ocular events while enabling a more natural level of control over the interface. The CNN and the EUDN replace the normally used feature-based ocular event detection methods for EOG. Finally, a Reinforcement Learning-based game actuation approach simultaneously updates multiple (State, Action) pairs for each rewarded outcome, intervenes to mitigate the consequences of wrongful game Commands, and can be used as part of a "shared-control" paradigm based on EOG. Results show a positive impact of Reinforcement Learning both in improving participants' game performance as well as in reducing some of their subjective workload indicators.

show abstract

Adaptive Robot Assisted Therapy Using Interactive Reinforcement Learning

Cited by 23 publications

References 21 publications

Task Engagement as Personalization Feedback for Socially-Assistive Robots and Cognitive Training

Task Engagement as Personalization Feedback for Socially-Assistive Robots and Cognitive Training

An Interactive Multisensing Framework for Personalized Human Robot Collaboration and Assistive Training Using Reinforcement Learning

A Reinforcement Learning Assisted Eye-Driven Computer Game Employing a Decision Tree-Based Approach and CNN Classification

Contact Info

Product

Resources

About