The innovations of Industry 4.0 have revolutionized the service sector, profoundly influ- encing daily life and potentially leading to the partial replacement of certain job roles with advanced robotic systems. However, integrating and reusing software capabilities in robotics remains a challenge, necessitating the creation of frameworks that enable reliable and adaptable applications. This thesis addresses the often overlooked human element in human-robot interaction models by introducing a groundbreaking model that simulates human decision-making processes, allowing robots to more effectively support workers in various environments. The thesis builds on existing research to offer a novel, expanding model of human behavior, which is based on brain science research, significant empiri- cal findings, and real-world applications with potential policy implications. The proposed model combines the proposer-predictor-actor-critic model with the neuroeconomics model, taking into account not only time constraints but also incorporating a reward mechanism for decision-making. This integration enables the model to more closely resemble hu- man behavior. Experiments were conducted to assess the probability of mission success within a specific time bound, and the critical factors contributing to the effectiveness of human-robot interactions were identified. The study primarily discusses the impact of four parameters on success rates and provides recommended ranges for each parameter.
Le innovazioni dell’Industria 4.0 hanno rivoluzionato il settore dei servizi, influenzando profondamente la vita quotidiana e portando potenzialmente alla sostituzione parziale di alcuni ruoli lavorativi con sistemi robotici avanzati. Tuttavia, l’integrazione e il riu- tilizzo delle capacità software nella robotica rimangono una sfida, rendendo necessaria la creazione di strutture che consentano applicazioni affidabili e adattabili. Questa tesi affronta l’elemento umano spesso trascurato nei modelli di interazione uomo-robot intro- ducendo un modello innovativo che simula i processi decisionali umani, permettendo ai robot di supportare più efficacemente i lavoratori in vari ambienti. La tesi si basa su ricerche esistenti per proporre un nuovo, più ricco modello del comportamento umano, basato su ricerche in neuroscienze, importanti risultati empirici e applicazioni reali con potenziali implicazioni a livello di policy. Il modello proposto combina il modello proposer- predictor-actor-critic con il modello neuroeconomico, tenendo conto non solo dei vincoli temporali, ma anche incorporando un meccanismo di ricompensa per il processo decision- ale. Questa integrazione consente al modello di assomigliare più da vicino al comporta- mento umano. Gli esperimenti sono stati condotti per valutare la probabilità di successo della missione entro un limite di tempo specifico, e sono stati identificati i fattori critici che contribuiscono all’efficacia delle interazioni uomo-robot. Lo studio discute principalmente l’impatto di quattro parametri sui tassi di successo e fornisce intervalli raccomandati per ciascun parametro.
Modeling human decision-making in interactive service robotic ap plications
LIU, JIACHENG
2022/2023
Abstract
The innovations of Industry 4.0 have revolutionized the service sector, profoundly influ- encing daily life and potentially leading to the partial replacement of certain job roles with advanced robotic systems. However, integrating and reusing software capabilities in robotics remains a challenge, necessitating the creation of frameworks that enable reliable and adaptable applications. This thesis addresses the often overlooked human element in human-robot interaction models by introducing a groundbreaking model that simulates human decision-making processes, allowing robots to more effectively support workers in various environments. The thesis builds on existing research to offer a novel, expanding model of human behavior, which is based on brain science research, significant empiri- cal findings, and real-world applications with potential policy implications. The proposed model combines the proposer-predictor-actor-critic model with the neuroeconomics model, taking into account not only time constraints but also incorporating a reward mechanism for decision-making. This integration enables the model to more closely resemble hu- man behavior. Experiments were conducted to assess the probability of mission success within a specific time bound, and the critical factors contributing to the effectiveness of human-robot interactions were identified. The study primarily discusses the impact of four parameters on success rates and provides recommended ranges for each parameter.File | Dimensione | Formato | |
---|---|---|---|
JIACHENGLIU_THEISE__Modeling Human Decision-Making in Interactive Service Robotic Ap- plications.pdf
accessibile in internet per tutti
Dimensione
4.5 MB
Formato
Adobe PDF
|
4.5 MB | Adobe PDF | Visualizza/Apri |
Executive_Summary_LIU_Modeling Human Decision-Making in Interactive Service Robotic Ap- plications.pdf
accessibile in internet per tutti
Dimensione
2.99 MB
Formato
Adobe PDF
|
2.99 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/209887