RT Dissertation/Thesis
T1 Autonomous Decision-Making based on Biological Adaptive Processes for Intelligent Social Robots
A1 Maroto Gómez, Marcos
AB The unceasing development of autonomous robots in many different scenarios drives anew revolution to improve our quality of life. Recent advances in human-robot interactionand machine learning extend robots to social scenarios, where these systems pretendto assist humans in diverse tasks. Thus, social robots are nowadays becoming real inmany applications like education, healthcare, entertainment, or assistance. Complexenvironments demand that social robots present adaptive mechanisms to overcomedifferent situations and successfully execute their tasks. Thus, considering the previousideas, making autonomous and appropriate decisions is essential to exhibit reasonablebehaviour and operate well in dynamic scenarios.Decision-making systems provide artificial agents with the capacity of makingdecisions about how to behave depending on input information from the environment.In the last decades, human decision-making has served researchers as an inspiration toendow robots with similar deliberation. Especially in social robotics, where people expectto interact with machines with human-like capabilities, biologically inspired decisionmakingsystems have demonstrated great potential and interest. Thereby, it is expectedthat these systems will continue providing a solid biological background and improve thenaturalness of the human-robot interaction, usability, and the acceptance of social robotsin the following years.This thesis presents a decision-making system for social robots acting in healthcare,entertainment, and assistance with autonomous behaviour. The system’s goal is toprovide robots with natural and fluid human-robot interaction during the realisation oftheir tasks. The decision-making system integrates into an already existing softwarearchitecture with different modules that manage human-robot interaction, perception,or expressiveness. Inside this architecture, the decision-making system decides whichbehaviour the robot has to execute after evaluating information received from differentmodules in the architecture. These modules provide structured data about plannedactivities, perceptions, and artificial biological processes that evolve with time that are thebasis for natural behaviour. The natural behaviour of the robot comes from the evolutionof biological variables that emulate biological processes occurring in humans. We alsopropose a Motivational model, a module that emulates biological processes in humans forgenerating an artificial physiological and psychological state that influences the robot’sdecision-making. These processes emulate the natural biological rhythms of the human organism to produce biologically inspired decisions that improve the naturalness exhibitedby the robot during human-robot interactions. The robot’s decisions also depend on whatthe robot perceives from the environment, planned events listed in the robot’s agenda, andthe unique features of the user interacting with the robot.The robot’s decisions depend on many internal and external factors that influence howthe robot behaves. Users are the most critical stimuli the robot perceives since they arethe cornerstone of interaction. Social robots have to focus on assisting people in theirdaily tasks, considering that each person has different features and preferences. Thus,a robot devised for social interaction has to adapt its decisions to people that aim atinteracting with it. The first step towards adapting to different users is identifying the userit interacts with. Then, it has to gather as much information as possible and personalisethe interaction. The information about each user has to be actively updated if necessarysince outdated information may lead the user to refuse the robot. Considering these facts,this work tackles the user adaptation in three different ways.• The robot incorporates user profiling methods to continuously gather informationfrom the user using direct and indirect feedback methods.• The robot has a Preference Learning System that predicts and adjusts the user’spreferences to the robot’s activities during the interaction.• An Action-based Learning System grounded on Reinforcement Learning isintroduced as the origin of motivated behaviour.The functionalities mentioned above define the inputs received by the decisionmakingsystem for adapting its behaviour. Our decision-making system has been designedfor being integrated into different robotic platforms due to its flexibility and modularity.Finally, we carried out several experiments to evaluate the architecture’s functionalitiesduring real human-robot interaction scenarios. In these experiments, we assessed:• How to endow social robots with adaptive affective mechanisms to overcomeinteraction limitations.• Active user profiling using face recognition and human-robot interaction.• A Preference Learning System we designed to predict and adapt the userpreferences towards the robot’s entertainment activities for adapting the interaction.• A Behaviour-based Reinforcement Learning System that allows the robot to learnthe effects of its actions to behave appropriately in each situation.• The biologically inspired robot behaviour using emulated biological processes andhow the robot creates social bonds with each user.• The robot’s expressiveness in affect (emotion and mood) and autonomic functionssuch as heart rate or blinking frequency.
YR 2022
FD 2022-03
LK https://hdl.handle.net/10016/35825
UL https://hdl.handle.net/10016/35825
LA eng
NO Mención Internacional en el título de doctor
DS e-Archivo
RD 16 jul. 2024