Rahardhita Widyatra Sudibyo

Battery management systems are essential in electric vehicles and renewable energy applications, especially in terms of ensuring optimal battery health and performance and regarding the state of charge (SOC) in batteries consisting of many cells. The lifetime and efficiency of the battery depend on the accuracy of the SOC parameter estimation. Moreover, systems that apply active balancing technology are able to move cells with high SOC data to cells with low SOC. Many methods have been developed, but their long execution time makes them less optimal when applied. High-speed SOC estimation is required in active balancing technology, in addition to high accuracy. Therefore, this study proposes the estimation of SOC parameters using a statistical and metaheuristic approach from voltage and current input data in each battery cell. The experimental results showed that the metaheuristic-based method (ANFIS) had better RSME and R² values compared with the polynomial and linear regression or even the machine learning-based method (recurrent neural network) for training data.

Battery management systems are essential in electric vehicles and renewable energy applications, especially in ensuring optimal battery health and performance. Especially regarding the state of charge (SOC) in batteries consisting of many cells. The lifetime and efficiency of battery depend on the accuracy of the SOC parameter estimation. Moreover, the system that applies active balancing technology will move cells that have high SOC data to cells that have low SOC. Many methods have been developed, but the long execution time makes it less than optimal when applied. The speed of SOC estimation is also required in active balancing technology in addition to the accuracy factor. Therefore, this study proposes to estimate SOC parameters using a statistical and metaheuristic approach method from voltage and current input data in each battery cell. The experimental results showed that the metaheuristic-based method (ANFIS) had better RSME and R2 values compared to the polynomial and linear regression method.

Maintaining water quality in fish ponds is crucial in aquaculture. Fish farmers strive to increase yields by improving water circulation, temperature, and pH levels in the ponds. Poor water circulation can lead to significant pH changes and disrupt the health and growth of the fish. In some areas, farmers use electrically powered water pumps for water circulation. For a more environmentally friendly solution, innovation in water pump design is needed. The proposed appropriate technology is a solar-powered water pump, which uses solar panels as the electricity source. Test results show that the solar panel can fully charge the battery from 7:00 AM to 5:00 PM. Testing of the DS18B20 temperature sensor shows good accuracy with a maximum error of 1.47%, while the pH sensor also works quite accurately with a maximum error of 1.8%, making them suitable for monitoring water quality in fish farming.

In smart city infrastructure, road management is a critical component of transportation infrastructure inspection to ensure a safe driving experience. The current method relies on a visual inspection or survey to monitor road conditions. Several studies have been widely developed to change visual surveys, such as ultrasonic detection systems, but the accuracy of the devices is limited due to environmental factors. LiDAR, on the other hand, provides a more accurate alternative by measuring distance. In this paper, we propose a low-cost infrared Time of Flight (ToF) sensor with a programmable Region of Interest (RoI) feature to adjust the pointing of the sensor. The RoI identifies road characteristics by setting a LiDAR array, including depth/height, width, and length information. The success of this system was tested through an experiment with an error rate of less than 20% for the width and depth measurement but 35% for the length measurement.

Proses kecepatan pencarian korban bencana alam merupakan faktor penting yang memengaruhi peluang korban untuk bertahan hidup. Sebagian besar lokasi yang terkena bencana alam akan sulit untuk diakses, dan saat ini tim penyelamat masih menggunakan alat berat untuk membuka akses tersebut. Hal ini membutuhkan waktu yang lama untuk menuju ke lokasi dan mencari korban. Pada penelitian ini, kami mengusulkan sebuah perangkat cerdas untuk membantu Badan Nasional Pencarian dan Pertolongan (BASARNAS) menggunakan drone yang dilengkapi dengan kamera untuk mencari korban bencana alam secara real-time. Dengan menggunakan perangkat cerdas yang dibuat ini, pencarian akan lebih efektif, karena ini dapat mempercepat tim penyelamat dalam mencari korban di lokasi bencana. Proses pendeteksian dan perhitungan jumlah korban dilakukan pada kamera menggunakan algoritma Convolutional Neural Network dan arsitektur You Only Look Once versi 4 (YOLOv4). Hasil dari penelitian ini adalah algoritma Convolutional Neural Network memiliki akurasi yang cukup tinggi yaitu sebesar 92,29% dalam mendeteksi kondisi korban. Selain itu tinggi kamera dan posisi dari kamera juga memengaruhi akurasi yang didapat, dimana hasil yang didapat dengan kamera statis memiliki akurasi yang cukup optimal yaitu sebesar 95%. Namun semakin tinggi kamera, akurasi akan semakin turun yaitu sebesar 67%.

The Internet of Things (IoT) has been extensively deployed for Smart Cities due to its ability to process many different and heterogeneous end systems. The IoT innovation encourages artificial intelligence applications to process data. In Smart City infrastructure, the road is a critical component of transportation infrastructure that supports the economic, social, and cultural things of community life and various aspects of community life. Road conditions affect a variety of community activities. Good roads enhance comfort and support local businesses. However, many roads remain in bad condition, such as potholes. Various methods have been attempted to identify potholes, especially the two-dimensional imaging method. This paper proposes the real-time Artificial Intelligence detection of potholes using the Convolutional Neural Network (CNN), which leverages the Edge Tensor Processing Unit (TPU) with the MobileNet SSD v2. The system was set up on a Jetson Nano with a few extras, including a camera and GPS, to support the IoT infrastructure. Evaluation for the model consists of device implementation, model evaluation, GPS position deviation, and on-road implementation. The effectiveness is confirmed through experiments using a system test-bed that generates ideal mAP off 0.22 and recall values.

Road is a transportation infrastructure that has a very important role in supporting the economic, social and culture as well as various other aspects of community life. Various community activities are affected by road conditions. Roads that are in good condition will provide comfort and facilitate economic activity in an area. However, in reality there are still many roads that are not in good condition, such as potholes. Various attempts have been made to detect potholes automatically, especially with the two-dimensional imaging method. In this paper, we propose real-time potholes detection using the Convolutional Neural Network (CNN) method based on the Edge Tensor Processing Unit (TPU) with the MobileNet SSD v2. Our system was implemented on Jetson Nano with several additional sensors such as a camera and GPS. The accuracy of the system is verified through experiments using the testbed.

In the IEEE802.11 wireless local-area network (WLAN), the TCP throughput unfairness will appear when multiple hosts concurrently communicate with a single access-point (AP), since the TCP window size, the modulation and coding scheme (MCS), and the packet drop rate become different among them. In this paper, we propose a TCP fairness control method using the transmission delay in WLAN. The target throughput is introduced as the equal throughput, which will be dynamically updated with measured throughputs. Then, the delay is controlled by the PI feedback control such that the measured throughput is equal to the target. The effectiveness is confirmed through experiments using our elastic WLAN system testbed with one AP and up to four hosts, where all the hosts have achieved the similar throughputs based upon the proposal.

For reducing the energy consumption of wireless local-area network (WLAN), we have studied the transmission power minimization of the access-points (AP). A throughput estimation model is adopted to estimate the minimum power satisfying the required performance after tuning the parameters by throughput measurement results with different powers in the network field. Then, the measurement minimization is essential to reduce labor costs in this approach. Thus, we proposed a method of minimizing host locations for measurements while maintaining the model accuracy. In this study, we present an improvement of this method by limiting measurements at two nearest locations from the AP. The experimental results for IEEE802.11n WLAN show that the number of measurements is reduced by 60% while keeping the throughput estimation error.