Based on computer vision technology, this research suggests an application to identify and assess a fitness practitioner’s movements. Several fitness movements such as lifting weights, squat jumps, and pull-ups that are very beneficial for health and body fitness become the main movement for body building. However, those kinds of activities may be very dangerous if done incorrectly. Based on the problem, we developed an application based on computer vision to recognize and correct the pose accuracy of fitness practitioners by using input in the form of videos that record the movements of fitness practitioners continuously. To categorize the many forms of fitness sport movements, this system uses the support vector machine (SVM) method. On the monitor screen, the classification results will be visible. The result shows that the accuracy of the system is 96.87% by using SVM with the Radial Basis Function (RBF) kernel type and can make corrections to four types of fitness movements with a testing accuracy of 90.62%.

Currently, much medical personnel died because of being infected by COVID-19 and because of low personal protective facilities and the duties of medical personnel that must carry out to deliver the logistics to patients and make many contacts between the medical personnel and patients of COVID-19. Mobile robots are considered the right solution to complete this problem. With mobile robots, hospitals or the place of isolation can minimize contact between medical personnel and patients of COVID-19 by carrying out the logistic delivery task. To deliver the logistic, a mobile robot must have low-level control, and the mechanism to carry out the workpiece also have the mechanism to open the door. The mechanism to carry out the workpiece is a system to pick up and place the rack of logistics from one place to another. In this study, the low-level control was applied using a PID control with the parameter's value $\boldsymbol{k}_{\boldsymbol{p}}=500,\boldsymbol{t}_{\boldsymbol{i}}=0.001$, and $\boldsymbol{t}_{\boldsymbol{d}}=0.001$ and teleoperation to control the mobile robot manually, so the mobile robot was able to move and carry out the load with the maximum value is 13 kg also open the door. Based on the results of the tests that have been carried out, the mobile robot with the proposed low-level control and the object management system can do the delivery task to reduce contact between medical personnel and patients of COVID-19, also the mobile robot can be controlled manually.

This paper presents a three-dimensional computer vision-based object recognition on FLoW-Vision in RoISC (formerly ER2C) has entered its second phase. Previously, the robot had a basic vision that was used to replicate ‘human-like’ visual skills using 2D computer vision. As a result of the above discussion, we proposed the design and implementation of an object recognition and pose estimation system based on three-dimensional computer vision to handle object recognition and pose estimation tasks in real-world environments simultaneously. In the object recognition process, a point-cloud segmentation method is used to obtain possible object clusters before starting the calculation of feature description. Then, a keypoints-based two-stage matching process is performed to speed up the computation of finding correspondences between the object clusters of the current scene and a colored point cloud model of an object. Next, a Hough voting algorithm is employed to filter out matching errors in the correspondence set and estimate the initial 3D pose of the object. Last process process the pose estimation from clustered object using RANSAC to search the largest surface as Z surface. Experimental validate the object recognition can work correctly with percentage 100% and pose estimation accuracy of the proposed system can work correctly with percentage 60% in a complex real-world scene.

In the last decade, the era of the Industrial Revolution 4.0 began. Technology played a major role in almost all industrial revolutions. So, computers and manufacturing systems will be able to collaborate. One of the pillars of the fourth Industrial Revolution is crowdsensing. Several experiments in the fields of society, infrastructure, agriculture, and health have used it. Crowdsensing in the health sector can help gather a substantial source of data on general public health conditions. However, most crowdsensing techniques rely on only one communication protocol. This method can cause problems if a system failure occurs. Besides that, another consideration is that the crowdsensing users consist of several communication protocols. As a result, we present a multi-communication protocol gateway architecture for crowdsensing that addresses the aforementioned issues by combining two communication protocols, such as HTTP and MQTT, into a single gateway. This gateway serves to receive data from sensor nodes before transmitting it to an IoT platform on a cloud server. The test results show that the MQTT protocol has better performance than the HTTP protocol in the delay test. The MQTT protocol delay reaches 0.113 seconds and the HTTP protocol delay is 0.717 seconds. In addition, the gateway is also able to receive data properly even though both protocols are run simultaneously.

Drone service is an Unmanned Aerial Vehicle (UAV) whose function is to deliver goods using air routes. Drone services sometimes have to travel long distances with a guaranteed security system to get to the customer's location and return safely to the base. But in fact, in the use of an insecure Internet of Drone (IoD) environment, there will be many attacks that try to manipulate the approximate location or interrupt transmission of data from fake nodes. Therefore, this paper focuses on a reliable data communication security system between drone services and server. Because the drone service uses a lightweight device, we propose a Hybrid cryptographic security scheme that has simple computing but a high level of security. It has layered encryption using AES- 256, ECC, and SHA256 which are provides high data authentication and encryption services for each node. The proposed scheme is efficient in terms of processing time so that it will not significantly affect the realtime sensor data from the drone. The experimental results show that the proposed system reaches 88.61 milliseconds to encrypt and decrypt messages using the Raspberry Fi device.

The study on balancing one-wheeled human transporter has previously done before in numerous studies. The main problem discussed in the study of one-wheeled human transporter is its balancing control system. The variation in the weight of the rider of the one-wheeled self-balancing human transporter becomes an important issue to the balance control system. The weight variation of the rider causes the response from the balance of the transporter become unstable. This paper will provide solution to solve these problems. In this paper a method of control system is presented therefore it provide robustness towards disturbance of rider's weight variation. The control method presented is the Proportional Integral Derivative-Active Force Control (PID-AFC). The PID-AFC method has the advantage of being able to provide robustness to disturbance in a system. This paper will seek for evidence that the PID-AFC can provide robustness towards disturbance, this method is compared to PID method. The experiment was carried out through a simulation with various rider's weights. After the experiment, it was found that the PID-AFC method is able to provide a better balance and provide robustness toward the disturbance of rider's weight variation on one-wheeled human transporters.

In human-robot interactions, robots that often interact with humans need some information from humans to be able to communicate. In human-oriented robots, the robots must pay attention wherever the human will go. Therefore, the human-oriented robots are more flexible. To estimate wherever the human will go, it can use the direction of the human or what is called intention. Intention can be taken from the perspective of the human body. An algorithm is needed to estimate the orientation of the human body. In this study, using image processing from depth image is expected to be able to estimate the orientation of the human body when a human is not moving (static orientation) so that it can be used by the robots to predict the intention direction of a human. One method that can be used to classify the orientation of a human body is the Support Vector Machine (SVM) so that it gets a classified orientation value. Then from the classified orientation value to get the static orientation value, Kalman Filter is used. Orientation classification with SVM method has accuracy of 96.528%. Static orientation estimation using this method produces precision value of 0.0970.

Technological innovations in the marine field have developed rapidly in the last decade. One of it is the innovation in the technologies related to the underwater research and exploration. It has been driven by the need of the a flexible, reconfigurable, and reliable machine for underwater operation termed Remotely Operated Vehicle (ROV). The ROV development has opened up many opportunities in the observation and exploration of underwater environment. The most important aspect in the ROV is the flexibility of motion during the maneuver in the water. The mechanical structure and also the static and dynamic analysis is very important aspect in designing ROV related to its operation environment. This paper proposes ROV body design based on analysis characteristic hydrodynamic and buoyancy effect using CFD. The characteristics of each fluid flow affect the motion of the ROV which movement is driven by the thrusters. The hydrodynamic characteristics and buoyancy effect will be analyzed. The result will be used in the further mathematical calculations in the controller design phase.

A network switch is an important network device on a computer network, the role of a switch can determine the good and bad of a network service. In a data center, sometimes it has several switch devices with different capabilities (diversity). For this reason, we need a strategy to carry out the effectiveness of using switch devices on computer networks. One way that can be done to realize this is by choosing the right partner between switches and hosts. The matching theory makes it possible to overcome this problem. In this study, it was simulated using the Stable Marriage Problem (SMP) method. The initial state to determine the ranking of pairs is done by using several attributes that are owned by each switch and host such as interfaces, services on the host, and data traffic on the device. Simulations are performed on mininet using Software-defined networking (SDN).

The three-dimensional approach in face identification technology had gained prominent significance as the state-of-the-art breakthrough due to its ability to address the currently developing issues of identification technology (illumination, deformation and pose variance). Consequently, this trend is also followed by rapid development of the three-dimensional face identification architectures in which some of them, namely Microsoft Kinect and Intel RealSense, have become somewhat today's standard because of its popularity. However, these architectures may not be the most accessible to all due to its limited customisation nature being a commercial product. This research aims to propose an architecture as an alternative to the pre-existing ones which allows user to fully customise the RGB-D data by involving open source components, and serving as a less power demanding architecture. The architecture integrates Microsoft LifeCam and Structure Sensor as the input components and other open source libraries which are OpenCV and Point Cloud Library (PCL). The result shows that the proposed architecture can successfully perform the intended tasks such as extracting face RGB-D data and selecting out region of interest in the face area.

This paper discusses about analysis the mechanical and forward kinematic of Prosthetic Robot Hand for T-FLoW 3.0, because it is the basic thing in building Prosthetic Robot Hand. This Prosthetic Robot Hand has 15 DoF which represented by 6 motors as the actuator. Mechanism which used for actuating each finger is by twisting the cable. The material used is PLA with 3D printed way for getting mass as light as possible and also easy to install. Mechanical analysis in this paper is to discuss the strength of the finger structure using the static structural analysis method. The kinematics that will be discussed in this paper is Forward Kinematic (FK). In mechanism design, FK is used to find constraint, calculate position vector of End of Efector (EoE). But, this paper only discusses the FK used to be looking for the constraint, it means that how maximum is the point that can be reached by finger with maximum degree of each joint of finger.

A map of a location made of a human mindset is made briefly by recording a situation already seen in memory. Human behavior is easy to do without a lot of thinking. The process is observed more deeply that when the human mind does not do the behavior in turn like looking at the whole room then just remember, but humans can see at once remember simultaneously. Bringing human behavior into a computer system is done by mimicking how humans look. Looking at using two cameras in combination with the sensor structure. To process the device requires a large computing. Be a challenge for Portable CPU that have lower capabilities than high end CPU. For that humanoid robot can perform tasks simultaneously and can accelerate, especially in image processing using portable CPU required a method that makes the processing into realtime. Achieving realtime with data from previous researcher can not be done with sequential algorithms when the complexity is high. After analyzing the problem was built parallel camera mapping process using multithread algorithm. This paper aims to accelerate the basic process of making the map so as to achieve realtime. This research aims to change the previous process that runs sequentially into a parallel system using multithread algorithm. The results of each process are compared between parallel systems with multithread algorithm and do not use parallel systems. There is a change in speed between the two processes. Further research is concerned with improving the mapping environment results when visualized.

Fire is one of the few disasters that frequently occurred in Indonesia. Jakarta has archived 1139 fire incident with more than 20 casualties in 2016. By the rapid development of information technology and communication in Indonesia, it turns out that the number of incident and casualty still cannot be suppressed. This research is focused on early fire detection and home monitoring based on Wireless Sensor Network using 4 sensors such as temperature, air humidity, Carbon Monoxide, and smoke sensors. If the fire probability calculation categorized as high, system will perform an action of unlocking home door lock, turning on alarm, and notifying the user. Based on the trials and testing result, system is able to send notification to user precisely. However, error still can be encountered with up to 0.3% ratio on fire probability calculation.

Pens-Wheel is an electric vehicle which uses one wheel that able to balance itself so the rider does not fall forward or backward while riding it. This vehicle uses one brushless DC motor as actuator which capable to rotate in both directions symmetrically. The vehicle uses a combination of accelerometer and gyroscope contained in IMU (Inertial Measurement Unit) for balancing sensor. The controlled motion on the vehicle occurs only on the x-axis (pitch angle), in forward and backward directions. The PID (Proportional Integral Derivative) control algorithm is used to maintain the balance and movement of the vehicle. From the simulation and application in the real vehicle, the use of PID control is capable driving the vehicle in maintaining the balance condition within ±10° tilt angle boundary on flat surface, bumpy road, and inclining road up to 15° slope.

The development of microkernel has sharply increased. One of the most successful microkernel implementation is L4. L4Linux is L4 version that is able to run virtualized Linux. We have also built our microkernel named FLoW. In this paper we described about our achievement in developing virtualized Linux on top of our FLoW microkernel. We implemented unique design about virtualizing more than one Linux. The idea is about creating more than one Linux to handle the requests made by user application about Linux system call. So every user application that requests the feature of Linux system call will be handled by more than one Linux Service. The result shows that there is an acceleration of processing time. The total CPU cycle used by user application in working with user request are less when having such mechanism. That means our mechanism successfully accelerate around 17% processing time of user application that request Linux system call from within our Linux Service that run on top of FLoW Microkernel. Then the system endurance testing resulted that our system is able to handle around 96 Linux system calls request almost at the same time without break.