A computational model of human decision making and learning for assessment of co-adaptation in neuro-adaptive human-robot interaction

Nowadays, many robotic applications require robots making their own decisions and adapting to different conditions and users. This work presents a biologically inspired decision making system, based on drives, motivations, wellbeing, and self-learning, that governs the behavior of the robot considering both internal and external circumstances. In this paper we state the biological foundations that drove the design of the system, as well as how it has been implemented in a real robot. Following a homeostatic approach, the ultimate goal of the robot is to keep its wellbeing as high as possible. In order to achieve this goal, our decision making system uses learning mechanisms to assess the best action to execute at any moment. Considering that the proposed system has been implemented in a real social robot, human-robot interaction is of paramount importance and the learned behaviors of the robot are oriented to foster the interactions with the user. The operation of the system is shown in a scenario where the robot Mini plays games with a user. In this context, we have included a robust user detection mechanism tailored for short distance interactions. After the learning phase, the robot has learned how to lead the user to interact with it in a natural way.

Download Full-text

Affective Goal and Task Selection for Social Robots

Handbook of Research on Synthetic Emotions and Sociable Robotics ◽

10.4018/978-1-60566-354-8.ch005 ◽

2009 ◽

pp. 74-87 ◽

Cited By ~ 17

Author(s):

Matthias Scheutz ◽

Paul Schermerhorn

Keyword(s):

Decision Making ◽

Real World ◽

Social Robots ◽

Human Robot Interaction ◽

Robot Interaction ◽

Task Selection ◽

Resource Limitations ◽

Effective Decision ◽

Effective Decision Making ◽

Selection Of

Effective decision-making under real-world conditions can be very difficult as purely rational methods of decision-making are often not feasible or applicable. Psychologists have long hypothesized that humans are able to cope with time and resource limitations by employing affective evaluations rather than rational ones. In this chapter, we present the distributed integrated affect cognition and reflection architecture DIARC for social robots intended for natural human-robot interaction and demonstrate the utility of its human-inspired affect mechanisms for the selection of tasks and goals. Specifically, we show that DIARC incorporates affect mechanisms throughout the architecture, which are based on “evaluation signals” generated in each architectural component to obtain quick and efficient estimates of the state of the component, and illustrate the operation and utility of these mechanisms with examples from human-robot interaction experiments.

Download Full-text

Human Decision-Making Behavior Modeling for Human Multi-Robot Interaction System

IFAC-PapersOnLine ◽

10.1016/j.ifacol.2020.12.2749 ◽

2020 ◽

Vol 53 (2) ◽

pp. 10202-10207

Author(s):

Wenhua Wu ◽

Jie Huang ◽

Zhenyi Zhang

Keyword(s):

Decision Making ◽

Behavior Modeling ◽

Robot Interaction ◽

Interaction System ◽

Human Decision ◽

Decision Making Behavior ◽

Multi Robot

Download Full-text

Brain Signals Classification Based on Fuzzy Lattice Reasoning

Mathematics ◽

10.3390/math9091063 ◽

2021 ◽

Vol 9 (9) ◽

pp. 1063

Author(s):

Eleni Vrochidou ◽

Chris Lytridis ◽

Christos Bazinas ◽

George A. Papakostas ◽

Hiroaki Wagatsuma ◽

...

Keyword(s):

Decision Making ◽

Ad Hoc ◽

Nearest Neighbor ◽

Human Robot Interaction ◽

Visual Signals ◽

K Nearest Neighbor ◽

Robot Interaction ◽

Brain Signals ◽

Fuzzy Lattice ◽

Fuzzy Order

Cyber-Physical System (CPS) applications including human-robot interaction call for automated reasoning for rational decision-making. In the latter context, typically, audio-visual signals are employed. Τhis work considers brain signals for emotion recognition towards an effective human-robot interaction. An ElectroEncephaloGraphy (EEG) signal here is represented by an Intervals’ Number (IN). An IN-based, optimizable parametric k Nearest Neighbor (kNN) classifier scheme for decision-making by fuzzy lattice reasoning (FLR) is proposed, where the conventional distance between two points is replaced by a fuzzy order function (σ) for reasoning-by-analogy. A main advantage of the employment of INs is that no ad hoc feature extraction is required since an IN may represent all-order data statistics, the latter are the features considered implicitly. Four different fuzzy order functions are employed in this work. Experimental results demonstrate comparably the good performance of the proposed techniques.

Download Full-text

Affective Goal and Task Selection for Social Robots

Social Computing ◽

10.4018/978-1-60566-984-7.ch140 ◽

2010 ◽

pp. 2150-2163

Author(s):

Matthias Scheutz ◽

Paul Schermerhorn

Keyword(s):

Decision Making ◽

Real World ◽

Social Robots ◽

Human Robot Interaction ◽

Robot Interaction ◽

Task Selection ◽

Resource Limitations ◽

Effective Decision ◽

Effective Decision Making ◽

Selection Of

Effective decision-making under real-world conditions can be very difficult as purely rational methods of decision-making are often not feasible or applicable. Psychologists have long hypothesized that humans are able to cope with time and resource limitations by employing affective evaluations rather than rational ones. In this chapter, we present the distributed integrated affect cognition and reflection architecture DIARC for social robots intended for natural human-robot interaction and demonstrate the utility of its human-inspired affect mechanisms for the selection of tasks and goals. Specifically, we show that DIARC incorporates affect mechanisms throughout the architecture, which are based on “evaluation signals” generated in each architectural component to obtain quick and efficient estimates of the state of the component, and illustrate the operation and utility of these mechanisms with examples from human-robot interaction experiments.

Download Full-text

Modeling and control of an active knee orthosis using a computational model of the musculoskeletal system

Journal of Mechatronics Engineering ◽

10.21439/jme.v1i3.19 ◽

2018 ◽

Vol 1 (3) ◽

pp. 12

Author(s):

Denis Mosconi ◽

Polyana Ferreira Nunes ◽

Adriano Almeida Gonçalves Siqueira

Keyword(s):

Computational Model ◽

Inverse Dynamics ◽

Human Robot Interaction ◽

Robotic Rehabilitation ◽

Robot Interaction ◽

Modeling And Control ◽

Interaction Control ◽

Knee Movement ◽

Feedforward Controller ◽

And Control

One-third of the stroke survivors remain with some disability, needing assistance to perform the activities of daily life and therapy to recover the lost functions. The robotic rehabilitation is a promissed field in this context improving the effectiveness of the treatment. Many researches have focused on developing human-robot interaction control to ensure user safety and therapy efficiency, but the validation of these controllers often requires contact between humans and robots, which involves cost, time and risk of accidents. This work aims to present a computational model of an ideal active orthosis used to assist the knee movement as a tool for test and validate human-robot interaction controls. Three controllers were applied to make the orthosis move the knee tracking the desired trajectory: a PID controller, an Inverse Dynamics-Based controller, and a Feedback-Feedforward Controller. The model proved to be useful and the controller with the best performance was the Feedback-Feedforward one.

Download Full-text

Developing an Educational and Research Human Robot Interaction Environment for a Mechanical Finger/Hand

Volume 4A: Dynamics, Vibration, and Control ◽

10.1115/imece2015-52305 ◽

2015 ◽

Author(s):

Christopher E. Ábrego ◽

Panos S. Shiakolas ◽

Michael R. Sobhy

Keyword(s):

Decision Making ◽

Success Rate ◽

Open Loop ◽

Human Robot Interaction ◽

Environmental Data ◽

Work Environments ◽

Post Processing ◽

Robot Interaction ◽

Threshold Test ◽

Human Operators

Human robot interaction (HRI) has become a growing area of study due to the increasing application of robots in work environments and daily life and interact and work in tandem with human operators. This synergy between humans and robots have expanded human endeavors and motivated neoteric and prospective areas of robotics research. The objective of this research is to develop an environment where both researchers and students can experience and experiment with different interaction modes with a robotic finger and hand. The acquisition and the post processing of environmental data from the hardware allow the user to envisage decision making algorithms. The present HRI environments have both closed and open loop controlled interaction modes for research and demonstration purposes. The interaction modes include direct open loop manipulation, force threshold interaction, voice commands, and variable force close. This additionally demonstrates the ease in functionality and augmentation in the testbed. As a result of speech recognition test, the testbed has a 79% success rate and has a 90% success rate in the force threshold test. The manuscript will also describe how certain actions are performed based on Prime States. An overview of the state algorithms are described as well as decision making diagrams.

Download Full-text