International Journal of Computing and Digital Systems ISSN (2210-142X) Int. J. Com. Dig. Sys. #, No.# (Mon-20..) Design, implementation, Interfacing and Control of Internet of Robot Things for Human Upper Limb Zeyad A. Karam 1, Mustafa Abdulkadhim Neamah 2 1 Department of systems Engineering, College of information engineering, Al-Nahrain University, Iraq 2Department of Computer Networks Engineering, College of information engineering, Al-Nahrain University, Iraq Received ## Mon. 20##, Revised ## Mon. 20##, Accepted ## Mon. 20##, Published ## Mon. 20## Abstract: Robotic arm plays a great role in assisting the human hand in current life. The robotic arm can be useful in many fields: such as industry, medicine and chemistry that demand more power, higher speed, and accuracy that cannot be provided by the human hands. This project involves the design and implementation of two degrees of freedom (DOF’s) assistant robot for aiding humans in the medical recovery of human upper left limb motion. The contribution in this work involves two parts: the first one is the use of microcontroller (Arduino card) for interfacing and processing the designed reading and writing circuits based on a designed MATLAB fuzzy PD controller and special Arduino Simulink library. The second contribution is adding a system's spring to the first link to rise the motion stability and increase the required torque in the first joint after the human hand attached to the robot links. The results illustrate the power of these circuits design and the error reduction occurred after supporting the system’s spring. An Internet of robotic thing (IoRT) model is designed, and then the robot arm is connected using our designed internet – based control system using IoT protocols for communication. Keywords: 2DoF's assistant robot, IoRT, Arduino board as interfacing card, Fuzzy controller. robot for the arm therapy application and the activity of 1. INTRODUCTION daily living with the name of 'Armin'. It has six DoF's, and The assistant robot is a program used for therapy for the it is equipped with position, force sensors and PID persons who have suffered from injuries in the nervous controller to fulfill the requirements of the prerecorded system, stroke, brain trauma and sports. International trajectory mode and the predefined motion therapy mode Federation of Robotics (IFR) defines a disability arm as a [2]. The drawback was in the workspace limitations in robot implemented for help injures who have physical motion with all human hand joints. In 2005, Kiguchi et al., disordered that impede life activities [1]. The area of developed a Neuro-Fuzzy controller for computing torque expertise that robots is known as "disability robotics". for each joint in assistant robot. The exoskeleton was used Disability arm used to help people who are reordered them to assist the human motion in daily activities [3]. Argall and limbs after strokes of injuries that affect their life tasks. Brenna present in 2018, the introduction of clinically Several types of research deal with modeling and design of feasible autonomy solutions for rehabilitation robots, and controlled upper limb assistant robots that are used in opportunities for autonomy within the rehabilitation human limb assisting program. Upper limb exoskeleton domain, where they presented the full required designs for robot is another name for the assisting robot, these robots upper or lower limbs and what the robot autonomy required are proposed with several designs.: these exoskeleton with each case whether wearable or non-wearable [4]. robots have been used as an assisting device, a human Aitziber et al, proposed in 2018 the effective sensors, amplifier or a haptic interface. Most of them have less than were used as a substitute for the actual force and motion 7 DoF's. The following research deal with upper limb sensors of the proposed assisting robot that used for upper assisting robots that were implemented previously. For limb recover. These virtual sensors established the required modeling or controlling assisting robot, many ideas were stimulates of force and motion at the contact points, where proposed based on the motion geometry and controller the person interacts with the assisting arm using the arm type. The illustrated literature present verity kinds of ideas mathematical model [5]. that used to model and control of many structures of In 2018, Monica, et al, provide a control structure used assisting arm: Nef and Riener in 2005, proposed a new the electromyography signals for drive the joint motion and E-mail:author’s email http://journals.uob.edu.bh 2 Author Name: Paper Title … control the assisting arm. The relation analysis for assisting robot. The mechanical design and implementation electromyography signal and the subject force exertion was is accomplished. The robot involves 1 DoF for shoulder noticed. The authors provide surface of electromyography motion and 1 DoF for elbow motion. Two interfacing force-based effective control structure that can control the circuits based on Arduino card will be designed and force exerted by the assisting arm through the training [6]. implemented; one for reading a DC motor's position from the shaft encoder sensor and another to drive the motors to In 2019, research concerning design rehabilitation the desired position using control signals. APD fuzzy type robots witnessed major breakthroughs. For example, 1 controller architecture in Matlab Simulink will be Mohammad Hossein et al, presented a under lab designed to control the robot motion according to the implemented robot and testing the accuracy by evaluation desired treatment trajectories via the connection with the of five structures for testing of human-assisting arm Arduino card interfacing simulation library. interaction torques. These structures involve the simplest idea of using direct control for motor torques with The other sections of this article are arranged as consideration of the arm dynamics, also involves advanced follows: section 2 discusses the IoRT paradigm, while ideas with full arm dynamics such as inverse dynamics [7]. section 3 deals with the mathematical model of the proposed 2 DoF's assisting robot. Section 4 highlights the Edwin Daniel, et al. present an ordered review of arm- proposed mechanical design and its implementation. based systems focused on upper human body rehabilitation, Section 5 explains the proposed interfacing circuits design where these systems adoption in clinical training remained in two parts: one for reading the motor position sensors and limited. To understand the reasons of these limitation's, the other for driving the motors based on the controller they created method for self-adaptation used for signals. The section also presents the interfacing with the personalizing the training, of injured–assisting arm Arduino card simulation library. Section 6 shows the interaction, where it contributes in training of movement design of the fuzzy type 1 controller and its simulation with generation factors [8]. robot kinematics, while section 7 aims at showing the Brock, et al. presented an authenticated investigation results of a simulated robot kinematics response and the based on machine vision with deep neural networks, where real-time robot position response for the implemented it's used for read the environment to assist the predictive robot. Section 8 explains IoRT implementation model. structure for control the robotic lower-limb.[9]. Finally, section 9 is dedicated to the conclusion of the design. Leiyu, et al developeda prototype of ankle rehabilitation assisting robot with three degrees of IEEE technical committee of automation and robotics freedom's, around a virtual fixed ankle joint as center. The defined the robotic system connected via network as a ankle center should harmonize with the virtual fixed center group of controlled devices that are linked together via through the rehabilitation training. Also, a full information wireless or wired communication network [14] and [15]. acquisition design was implemented to provide the human- Also, robotic networked applications can be one of the machine interactive for the assisting arm, patients, and two cases: therapists [10]. It can either be ‘teleoperated’, which means they are In 2014 and 2015, another creative design was established by Hassan and Zeyad, where they designed a 3 remotely controlled by a person from a network to achieve Dof's assisting robot arm for upper left limb to assist the the given task. One example of such systems is the Mars injured persons who suffer from losing control in their rover robot. Its picture shown in fig. 1. below: arms. The designed robot is a wearable one and it has a superior mechanism for adjusting the robot with multi human arms lengths, also it interfaced with PC using Advantech card via specially designed controlling circuits. The presented robot was managed by using a novel control strategy of force position controlling algorithm that incorporated a fuzzy type one as a position controller. The controller parameters were optimized using a modified particle swarm optimizer PSO. The presented robot was tested with multi human hand weights and entered the phase of injured person training and showed superior results in the execution of the medical trajectories [11], [12] and [13]. This work includes the design and implementation of 2 DoF's exoskeleton robot for assisting Figure 1. Mars rover robot. and training the human upper left limb as a non-wearable http://journals.uob.edu.bh Int. J. Com. Dig. Sys. #, No.#, ..-.. (Mon-20..) 3 five distinct layers of operation. A brief description for each layer will be given below: a. The hardware layer: These are the sensors, smartphones, vehicles, actuators, and other physical real-life components. b. The network layer: In this layer cellular connectivity and other communication and networking technologies are described, such as 3G, 4G, WIFI, NFC and LORA [17]. c. The internet layer: In this layer specific IoT protocols are added such as MQTT, XMPP and LLAP, to name a few.