Biomedical robotics is a branch of robotics that deals with the design and development of robots and robotic tools for use in the healthcare sector. It is a multidisciplinary research field in which robotics, medicine, computer science converge. Biomedical robotics spans different domains: surgery, diagnosis, rehabilitation, prosthetics, and assistance. The aim of this thesis is to design a robot capable of challenging an opponent on a board game, for rehabilitation purposes. The idea is to keep the patient busy during rehabilitation hours while allowing him to learn or improve mobility and strategic skills using a recreational game. To maximize the effectiveness of the therapy, the user experience must be as close as possible to a real game (e.g., chess or draughts) with a physical opponent. For this reason, the robot has to move the pieces on the board, mimicking the role of the other player. The design and development of the robot were performed in a simulated environment (Gazebo). The simulated robot is of the parallel type, having the end effector directly con- nected to the base through rigid structures that work in coordination. ROS (robot operating system) was used for the development and design of the robot controller, leveraging codes written in Python and C++. The first part of the thesis concerns the assembly of the physical structure of the robot through the Unified Robot Description Format (URDF). Secondly, the controllers for movement along the Cartesian axes have been included. In this case, it was im- portant to delimit the area of action that the controllers had on the various parts to cover the whole chessboard and the control strategies to move the pieces according to their type. Finally, the game intelligence was introduced, in order to enable the robot to autonomously play a chess game against the user. While a wide range of game in- telligence systems has been proposed in the years, medium-level intelligence has been used. In conclusion, we have implemented the simulated version of an autonomous robotic board game that could be translated to a physical system able to engage the user for rehabilitation purposes.
La robotica biomedica è un ramo della robotica che si occupa dello sviluppo di strumenti utili nel campo medico. Essendo un campo multidisciplinare, include diversi settori, tra cui chirurgia, diagnosi, robotica riabilitativa, prostetica, assistenza. L’obiettivo di questa tesi è quello di simulare un robot in grado di giocare a scacchi. Lo scopo riabilitativo è quello di mantenere impegnato il paziente durante la fase di recupero e aiutarlo nel recupero della mobilit`a muscolare. La scelta del robot è ricaduta su un dispositivo parallelo, avente l’end effector direttamente connesso alla base tramite strutture rigide. Per completare le simulazioni è stato usato ROS (Robot Operating System), che incorpora linguaggi di programmazione come Python e C++. L’ambiente di simulazione usato è Gazebo. La prima parte dell’elaborato riguarda l’assemblaggio delle parti tramite URDF, che utilizza il linguaggio dichiarativo XML per descrivere parti, giunture del robot e i potenziali gradi di libertà. Secondariamente sono stati inclusi controllori per il movimento lungo gli assi cartesiani. In questo caso è necessario delimitare l’area di utilizzo per considerare solo la scacchiera. Infine è stata adattata un’intelligenza di gioco di media forza per permettere ad utenti con poca esperienza di divertirsi contro il robot. Un’interfaccia grafica è stata incorporata per permettere al giocatore di intuire meglio quale pezzo stesse usando. In conclusione, trasportando un pezzo nell’interfaccia e aspettando alcuni secondi si riceve la risposta da parte del robot.
Robot-assisted game therapy : a case study in virtual reality
Schinelli, Roberto
2020/2021
Abstract
Biomedical robotics is a branch of robotics that deals with the design and development of robots and robotic tools for use in the healthcare sector. It is a multidisciplinary research field in which robotics, medicine, computer science converge. Biomedical robotics spans different domains: surgery, diagnosis, rehabilitation, prosthetics, and assistance. The aim of this thesis is to design a robot capable of challenging an opponent on a board game, for rehabilitation purposes. The idea is to keep the patient busy during rehabilitation hours while allowing him to learn or improve mobility and strategic skills using a recreational game. To maximize the effectiveness of the therapy, the user experience must be as close as possible to a real game (e.g., chess or draughts) with a physical opponent. For this reason, the robot has to move the pieces on the board, mimicking the role of the other player. The design and development of the robot were performed in a simulated environment (Gazebo). The simulated robot is of the parallel type, having the end effector directly con- nected to the base through rigid structures that work in coordination. ROS (robot operating system) was used for the development and design of the robot controller, leveraging codes written in Python and C++. The first part of the thesis concerns the assembly of the physical structure of the robot through the Unified Robot Description Format (URDF). Secondly, the controllers for movement along the Cartesian axes have been included. In this case, it was im- portant to delimit the area of action that the controllers had on the various parts to cover the whole chessboard and the control strategies to move the pieces according to their type. Finally, the game intelligence was introduced, in order to enable the robot to autonomously play a chess game against the user. While a wide range of game in- telligence systems has been proposed in the years, medium-level intelligence has been used. In conclusion, we have implemented the simulated version of an autonomous robotic board game that could be translated to a physical system able to engage the user for rehabilitation purposes.| File | Dimensione | Formato | |
|---|---|---|---|
|
TESI_MAGISTRALE_ROBERTO_SCHINELLI__.pdf
non accessibile
Dimensione
4.39 MB
Formato
Adobe PDF
|
4.39 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/10589/183839