The Unmanned Ground Vehicle (UGV) Robot Arm is needed to take samples of volcanic material, namely material samples in the form of rocks from the Volcano which will later be studied more accurately. The focus of this research is on developing a robotic arm to take samples of volcanic material mounted on the UGV. The Robot Arm consists of 3-DOF (Degree Of Freedom) equipped with a Gripper to make it easier to take samples of the material from the Top of the Volcano. The Robot Arm is placed on the upper front of UGV Robot body. The movement of the Robot Arm utilizes Rotational Motion in each DOF. The Kinematic Model of this 3-DOF robotic arm was built using the Denavit Hartenberg method. This study discusses the movement of the robot arm with 3 degrees of freedom where the robot used has one kinematic strand consisting of 3 joints with the revolute type. The angle at each joint can be determined using a method used in this research, namely forward kinematics and inverse kinematics. Experimental activity shows that the error occurs because the mechanical system is less precise rather than the method used

In this paper, the development of inspection code generator for hole cylindrical feature evaluation is reported. The development process aims to enable on-machine measurement process through Computer Aided Inspection Planning (CAIP) software. A CNC machine equipped with touch probe measuring tool is utilized in this research as an object for On-Machine Measurement (OMM) System. The hole feature as measurement object were defined from feature recognition process. An algorithm to construct inspection path planning was developed according to hole feature measurement point. Inspection path planning define touching point, retract point, clearance ratio, and retract distance from the measurement point. The inspection path planning is transformed into inspection code based on Heidenhain iTNC 30i controller code standard. The Inspection code were experimentally executed in the DMG 50U CNC Machine to validate the path planning. Renishaw touch probe were used to measure the hole feature dimension and position. The OMM result was compared to the vernier and measurement result. The experiment result shows that the proposed method has accurate equal inspection result to the vernier.

Kontes Robot Pemadam Api Indonesia (KRPAI) is a division of the Robot Competition in Indonesia whose mission is to extinguish fires according to the rules of the Trinity College International Fire Fighting Robot Contest (TCIFFRC) in Hartford, United States. The robot is placed in one of the 4 rooms randomly and is required to find a fire then extinguish it and return to the room where the robot is placed. The ability of robots to be able to carry out missions perfectly often fails because of the disturbance of obstacles in the room, do not know the pose of the robot when outside the room. A map is very important so that the robot knows its position in space and the distance of the robot from obstacles. Therefore, the focus of this research is not only on simulations but also on direct map-making to dynamically moving four-legged robots. The odometry method on the four-legged robot is used for routing and localization algorithms in making maps. Bresenham's algorithm is implemented in the map creation process. The results of the test and analysis show that a 1 mm grid map to a 5 mm grid map can describe parabolic-shaped obstacles, and the average error cell value for corridors, door widths, and walls is 0.92 cells.

This work presents a fast and simple learning algorithm for humanoid robot walking gait cases. The standard method of reinforcement learning takes too much time to learn a stable walking gait. Thus, we propose a rule-based learning method that has never been used in this kind of walking gait learning case. We implement our method in a simplified TFLoW humanoid robot model in simulation software CoppeliaSim. The result shows by using our proposed method, T-FLoW humanoid robot can walk for 200 steps after taking the learning process for about 800 episodes and has a better walking performance than the classical pattern generation for planning a walking gait motion.

The manipulator robot on the humanoid robot has functioned as an arm to grasp objects. The end-effector position of the robot is must first be known to perform the grasping task. Therefore, using the kinematics solution to find the robot end-effector position in the Cartesian space. This research paper presents the inverse kinematics of the 7-DOF T-FLoW humanoid robot manipulator using the Improved Damped Least Squares method with joint limits to avoid mechanical limitations. Forward Kinematics with the Homogeneous Transformation Matrix is used in the solution to find the current position of the end-effector in the Cartesian space. This research using the DLS method because it can avoid kinematic singularities and performs better than pseudoinverse based formulations. The experiment results show that the improved solution is more robust in performing joint limitation with a success rate of 100% and generating more natural motion than the original DLS.

Detecting dynamic and static objects is one of the important abilities for a mobile robot, including a healthcare mobile robot. As long as the robot carries out its duties, it will often encounter objects, both dynamic and static. Therefore, recognizing the dynamic and static objects is crucial. In this paper, the existence of human as an example of dynamic objects is obtained using a Laser Range Finder (LRF), since it can work faster than ordinary cameras. To recognize human data obtained from LRF, a sliding window process is applied to get the signal data of human feet which will then be classified using the Support Vector Machine (SVM). Meanwhile, to overcome the difference in the size of the human foot signal caused by the distance changes, a pyramid scanning process is also applied. Based on the experimental results obtained, human can be detected from a range of distances 1 up to 4.25 meter within 73 msec.

Drive technology for mobile robots is currently developing very quickly, especially for the type of wheeled driven platform. The driving models such as Ackermann steering, DDMR, and Omni-wheel robots have been widely implemented as mobile robot platforms. However, to answer the challenges in the robot contest where the competition is getting tougher, research is needed on a mobile platform that is more efficient, faster, and more precise. In this research, the design and fabrication of a mobile robot platform with a swerve drive model will be carried out. Swerve drive has independent driving and steering at each wheel. This model can provide a higher speed and freedom of maneuver for the robot compared to the DDMR, Ackermann steering, and Omni wheel drive models. However, swerve drive requires a higher number of motors as well as a more complex control algorithm in regulating the speed of the wheel drive motor and the steering angle on each independent wheel to be further controlled simultaneously in moving the robot to the target position and orientation. In terms of the control system, a multi-level control will be made where the low-level control will regulate the speed of the wheel drive and its relative direction to the robot’s body. Meanwhile, high-level control will be used to coordinate the movement of each wheel so that the results can make the robot to move according to the given trajectory.

Nowadays, computer-based tests are commonly used since a pandemic made educational institutions hold exams from home and it made computer-based tests application demand increasing time by time. This situation leads to the need for applications that can run on any device, secure from cheating and compatible with poor internet connection. Our proposed system able to improve security and deception issues in the examination. In this paper we also introduce distributed online computer-based test for English test that made our software can share questions-data to another instance securely.

This paper proposes an adaptive control strategy to improve the performance of the Social Force Model (SFM) based mobile soccer robot navigation using the Fuzzy Inference System (FIS). The combination of FIS dan SFM is then called a Fuzzy Social Force Model (FSFM). In the FSFM strategy, FIS is used to adaptively adjust the parameters of SFM based on the stimulus from the external condition, namely the obstacle's relative distance to the robot, d, and its direction, γ, so that the robot's reactivity and responsiveness can be automatically adjusted. Currently, we adjust only one parameter, namely obstacle's gain value, k. Obstacle's gain value will control the amount of forces produced by the closest obstacle. We tested our proposed method using a realistic 3D simulator, called V-Rep, by utilizing an omnidirectional robot model from Festo, namely Robotino. Our experimental results show that our proposed FSFM can work well by always successfully finishing all of the trials with less collision with obstacles. The comparison results with the fixed-parameter method prove that our proposed method is better and very promising to be implemented for real robot applications.

E-learning is distance learning that uses computer technology, networks of computers and the internet. E-Learning allows students to study via computers in their respective places without having to go to study/lectures in class physically. Moodle is a Learning Management System that is used as a medium for delivering E-Learning. The problem that often arises in e-learning is that in the learning process, students interact more with e-learning media so that teachers will find it difficult to monitor student behavior when using learning media. In fact, students in some cases tend to drop out or attend lesser classes. Moodle can capture student interactions and activities while studying online using log files. From the results of student interactions and activities on e-learning, it can be used to determine their learning style. Identifying student learning styles can improve the performance of the learning process. This research suggests an approach to automatically predicting learning styles based on the Felder and Silverman learning style (FSLSM) model using the Decision Tree algorithm and the ensemble Gradient Boosted Tree method. We've used actual data sets derived from e-learning program log files to perform our work. We use precision and accuracy to assess the results. The results show that our approach is delivering excellent results.

This study focuses on the feature extraction and development of tomato growth model. In Indonesia, the application of sensors and actuators to build smart farming has become increasingly popular in the past year but to produce measured harvest quality and quantity in smart farming, a model of growth and yield of the tomato plant is needed. in this paper, observations on the growth of tomato plants were carried out to extract features in the growth of tomato plants. In the experiment, a greenhouse was built to maintain the growth conditions of tomato plants which are equipped with an information system for reading sensor data to determine the greenhouse environment conditions for the growth of tomato plants with a local-global management strategy. from the experiment, we obtained a response model of the tomato plant growth with the effect of the maximum temperature that caused the blossoms to fall. and from the growth model, there are 5 growth stages of tomato plants are obtained, which at each stage has different growth variables with different nutritional needs and environmental conditions. with the important knowledge gained from this experiment, it can be used to develop a dynamic tomato plant growth model based on the growth stage, which is expected to be applied in the control system on smart farming in the future work.

Recently, 3D visualization technologies such as point cloud are growing rapidly and are in demand for environmental mapping needs because they have the ability to represent the real world in a more complete and in depth way, as well as new and more advanced visual experiences. The point cloud model is able to represent objects well through a series of points and is increasingly used in various fields of research. While 3D visualization technology is growing, several platforms such as humanoid robots, mobile robots, and automatic cars require mapping technology that is able to properly map the surrounding environment with a better display to improve intelligence and decision-making abilities. To overcome the problem of mapping, in this research an experiment was conducted to combine point cloud to form a map based on a visual odometry from stereo camera.

This paper discusses about research for developing human-like robots. To make the robot walk like a human on inclined surface, it is currently being developed into two major parts, namely the mechanical system and the equilibrium motion control system. In this mechanical system, it discusses the relationship of all human parts that are modeled on each joint and link to the robot. The analysis used in modeling this mechanical system is dynamic walking and Dynamic Step for robot on the inclined surface . Whereas for motion control systems on inclined surface, Online Motion Stabilization and Online Posture Stabilization analysis is used. Planning balance movements is used for each joint to make the robot walk stable in inclined surface. Balance system planning is made based on input data obtained from sensor readings from the position of the robot. Then the triple inverse pendulum approach can be used which can be solved by simulating the movement from the feet and body . The Walking Simulation Result have success for dt = 20 ms. It means that the distances and joint values won`t be updated any faster that every 20 ms .

The defense on food today is mostly depend on agriculture. However, it is predicted that on 2050, the food production from agriculture will reduce and be replaced by aquaculture. Indonesia may take the advantage from this situation. Since it has the second longest of its coastline in the world, Indonesia has a potential to be number one for fishery production from aquaculture field. Therefore, survival rate of the creature inside the pond must be very high. One of the factors that influence survival rate of creature in the pond is water quality, which is influenced by physical, biological and chemical properties. In this study, we deal with biological content in the pond, of which several bio-organism live in. Four type of planktons are taken as samples i.e., Chlorella Vulgaris, Thalasiosira Sp., Skeletonema Sp. and Skeletonema Costatum. Several plankton contents, 0%. 5%, 10%, 25% and 50% are prepared. This sample is then observed under digital microscope and RGB LED. The result from digital microscope and RGB LED are then plotted and taken into formulation model. From this model, four types of plankton are possible to be classified.

This paper discusses how to monitor the operation of guest OS on the virtualization environment. There are several forensic applications which are useful for analyzing events that run on operating systems specifically memory, CPU, and I/O. We use the LibVMI library to create a new plugin for Drakvuf in order to detect Hang on a guest OS that runs on top of the Xen Hypervisor. The experiment shows that libVMI able to monitor the state of the guest OS. However, the a problem arises when LibVMI start to detect the hang condition. The EPT violation occurs during hang monitoring process.