23 February 2023 in Jurnal Teknologi Informasi dan Multimedia
This research focuses on the load and stability of unmanned aerial vehicles that are as light as possible but with optimal stability functions and increasing the capacity of liquid fertilizer lifted by unmanned aerial vehicles. The increase in lift is related to the payload and stability to be carried by the unmanned aerial vehicle. By optimizing the arm using the arm carbon method, so without reducing the strength and function of the drone, the gross weight of the components becomes much lighter, so that the weight can be focused on increasing the capacity of liquid fertilizer. With the increase in liquid fertilizer capacity, of course, the weight also increases, this raises a new problem, which is to make the balance value when the drone take-off has a large magnitude, thus affecting stability which causes the potential to fail to fly, therefore this research was carried out so that the drone has stability. maximum and can minimize the effects of shocks from increasing capacity after using arm-carbon. The measurements taken were the success rate of the system in flying over agricultural land, power testing and current measurements. After testing, it was found that by using a split tank and carbon arm system, current and power tend to be more stable, so that battery life is longer, liquid fertilizer capacity can be increased and the potential for flight failure can be minimized with a split tank system with 90% accuracy, carried out optimizing weight with carbon arms and increasing the capacity of liquid fertilizer by 30%, using a split tank system increases the stability of the drone at take-off by up to 80%.
23 July 2025 in International Journal of Technology, Food and Agriculture
This study aims to develop a transporter robot as a learning tool for elementary and secondary school students to introduce them to robotics. The robot features a differential drive system and a lifting arm akin to a forklift system. The robot is controlled using an ESP32 microcontroller and connected to an Android application through a Firebase cloud database. Utilizing an IoT platform, the training can be conducted online, allowing students to control the robot remotely while it remains at the training location. The transporter robot is designed to provide an interactive and hands-on learning experience in robotics, enabling students to grasp fundamental concepts in robotics, programming, and control systems. The differential drive system offers excellent maneuverability, while the lifting arm allows the robot to perform tasks such as lifting and moving objects, adding functional aspects to the learning process. Test results indicate that the robot performs well under online control, with fast and accurate responses to commands from the Android application. The use of Firebase ensures stable and real-time communication between the robot and the application. This implementation not only enhances the accessibility of robotics training but also enables the development of students' STEM skills in a more flexible and affordable manner. The study concludes that an ESP32-based transporter robot with online control via an Android application and Firebase cloud database can be an effective and innovative learning tool. This tool aims to enhance students’ STEM skills -especially in programming, control systems, and mechanical design- through interactive, real-world applications.
28 May 2020 in Jurnal Informatika Polinema
Tingkat dehidrasi tidak hanya bisa dirasakan secara langsung akan tetapi dapat diamati dan dilihat secara fisik berbasis visual. Secara visual salah satu gejala dari dehidrasi dapat dilihat dari warna urine. Gejala ini biasanya tidak begitu diperhatikan dan dianggap biasa. Padahal gejala hipohidrasi atau dehidrasi merupakan dampak yang merugikan dari asupan air yang tidak memadai sehingga mempengaruhi warna urine yang dihasilkan. Kesulitan panca indra manusia membedakan gejala dehidrasi dan melihat perbedaan warna urine secara visual sering diterjemahkan berbeda-beda, dikarenakan tingkat kemiripan warna yang dihasilkan. Beberapa penelitian menunjukkan adanya pemanfaatan teknologi kamera dengan sistem cerdas dapat membantu kesulitan dan keterbatasan panca indra manusia. Penelitian ini menggunakan citra urine diambil dari sample orang dewasa yang dikelompokkan berdasarkan kategori warna urine hasil penelitian terdahulu. Pengambilan fitur dari setiap citra urine diambil nilai warna dari YCbCr. Model warna yang dihasilkan dari setiap sampel akan diidentifikasi menggunakan algoritma Random Forest dengan cross-validation. Hasil dari percobaan yang dilakukan menunjukkan akurasi 90% dari 30 dataset yang diujikan dengan nilai precision 90.2%, recall 90%, Mean absolute error 0.2473, dan Root mean squared error sebesar 0.3208.