Claim Missing Document
Check
Articles

Found 5 Documents
Search
Journal : International Journal of Aviation Science and Engineering

Bird Detection System Design at The Airport Using Artificial Intelligence Ummah, Khairul; Hidayat, Muhammad Fadly; Kurniawan, Denni; Zulhanif, Zulhanif; Sembiring, Javensius
International Journal of Aviation Science and Engineering - AVIA Vol. 4, No. 2 (December 2022)
Publisher : FTMD Institut Teknologi Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47355/avia.v4i2.72

Abstract

Bird strike is a process of crashing between bird and airplane which occurs in flight phase. Based on data, there are 40 times bird strike occurs every day (FAA, 2019). There are lot of research that already conducted to decrease number of birds at the airport. But it is not given significant changes. Hence, it is needed a model that can detect bird at the airport so that we can decrease the number of birds. Study already conducted by comparing motion detection with object detection and filter which can be used to improve detection quality. Model already developed using YOLOv4 object detection with 71.89% mean average precision. It is expected that object detection can be developed to become a bird repellent system in the future
Deep Learning Implementation on Aerial Flood Victim Detection System Ummah, Khairul; Hidayat, M Thariq; Risano, A Yudi Eka
International Journal of Aviation Science and Engineering - AVIA Vol. 4, No. 2 (December 2022)
Publisher : FTMD Institut Teknologi Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47355/avia.v4i2.73

Abstract

Hydrometeorological hazard such as floods are considered as a regular natural disaster in Indonesia due to its frequent occurrence. To mitigate the risk, search and rescue operations need to be carried out immediately. The sheer magnitude of floods poses a major challenge for responders, and the emerging drone technology could help to alleviate the problem due to its deployment speed and coverage. Automation in drone technology has potential to improve its effectiveness. This paper explores the idea of human detection during floods using a computer vision approach. This approach utilizes a one stage detector model as detection speed is crucial in disaster management case. The dataset used for training consists of 200 labelled and negative images taken from drone point of view. This paper conducted 3 experiments to find out the difference in performance when the model was trained on flood and non-flood dataset, as well as the effect of image input size to the model’s performance. The first experiment was trained on non-flood dataset. The second experiment was trained on flood dataset, and the third experiment is the modified version of the second model. The results show that the model trained on flood dataset performed worse than non-flood counterparts with the non-flood mAP reached 90.80% while flood mAP reached 39.15%. In addition, the experiments also conclude that increasing the input size of image during training, will increase the detection performance of the model at the cost of FPS
Development of Anti-UAV System Using Visual Artificial Intelligence Sembiring, Javen; Prianggoro, Dimas; Saputra, Rizal Adi; Tarkono, Tarkono; Ummah, Khairul
International Journal of Aviation Science and Engineering - AVIA Vol. 5 No. 1: (June 2023)
Publisher : FTMD Institut Teknologi Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47355/avia.v5i1.83

Abstract

Unmanned Aerial Vehicles (UAV) was first developed as a tool for military purposes. Due to the rapid growth in technology, UAVs are now used in various applications including civil needs. Of course, there are consequences for this where UAVs can be misused by irresponsible parties. One example is the use of UAVs in airport fields which can disrupt the airport operations and possibly become a serious threat towards security and safety of flights in the airport. This paper will discuss the artificial intelligence (AI) modeling to detect UAVs. This AI modeling is the first step in designing counter unmanned aerial system (C-UAS). UAV detection will use deep learning using YOLOv4 (single-stage detection) for optimal detection speed and accuracy. There are a total of 500 image data processed and used in two AI modeling experiments in this study. Gaussian blur filter is used to create dataset variations so that the training can be processed more efficiently and the model can detect better. The results shows that the training dataset that has been processed with gaussian blur (filtered dataset) increases the AI model’s detection performance in rainy and clear conditions. Therefore, the model trained using filtered datasets is more suitable for use in detecting UAV objects in anti-UAV systems.
Aircraft Detection in Low Visibility Condition Using Artificial Intelligence Ummah, Khairul; Widyosekti, M. Dhiku; Arif, Yanuar Zulardiansyah; Saputra, Rizal Adi; Riszal, Akhmad; Sembiring, Javensius
International Journal of Aviation Science and Engineering - AVIA Vol. 5 No. 1: (June 2023)
Publisher : FTMD Institut Teknologi Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47355/avia.v5i1.84

Abstract

Bad weather often interferes with the functioning of the air transport system. One example is the frequent flight delays for commercial aircraft, resulting in losses for both the airline and passengers. Artificial Intelligence (AI) technology can now minimize delays caused by bad weather, especially in low visibility conditions. This paper discusses AI modeling that can detect aircraft in a low visibility weather condition, especially in the airport area. The employed method is the deep learning approach with the YOLOv4 algorithm (single-stage detection), which is regarded as one of the optimal platforms in this field. There are 600 images used in this work to create and train three different models. Image Dehazing filter is employed on the training data before it is trained to produce the detection model. The result shows that the model has a good performance in terms of performance metrices. Thus, this model is suitable to be used to detect aircraft in low visibility conditions.
Clustering and BiLSTM Network for Aircraft Trajectory Prediction Model Sembiring, Javen; Fauzan, M Ariq; Ummah, Khairul; Hamdani, Fadil; Djuansjah, Joy R P
International Journal of Aviation Science and Engineering - AVIA Vol. 5 No. 2: (December,2023)
Publisher : FTMD Institut Teknologi Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47355/avia.v5i2.89

Abstract

The increasing demand for air travel requires the development of more accurate aircraft trajectory prediction methods to optimize airspace utilization and enhance safety. This paper presents a hybrid approach for single-flight-route trajectory prediction that employs the K-means clustering and Bidirectional Long Short-Term Memory (BiLSTM) networks. The primary objective is to develop a deep learning model that effectively predicts aircraft trajectories. Additionally, this research investigates the influence of trajectory clustering on prediction accuracy. To fulfill the objectives, a four-step methodology: data preprocessing, model construction, validation testing, and analysis is employed. Real-world historical flight data is used to train the BiLSTM model after being clustered with K-means. The model's performance is evaluated using randomized enroute flight data and various metrics like mean squared error and root mean squared error. This research is successful in accurately predicting the flight and the clustering process was proven to increase prediction accuracy by 15 percent in latitude, and 10 percent in longitude.