Research Projects
IRoS lab focuses on developing intelligent systems for robotics, biomedical, and surveillance. The general roadmap and track record are shown in the following page.
IS for Robotics (IS4Robotics)
Swarm Drone
Wisnu Jatmiko, Ario Yudo Husodo, Aprinaldi Jasa Mantau, et al.
Explores the development and simulation of coordinated drone operations using AirSim, an advanced physics-based platform. The project focuses on designing, testing, and optimizing multi-drone behaviors for tasks such as search-and-rescue, mapping, and environmental monitoring. Leveraging AirSim's simulation capabilities, the study aims to improve swarm intelligence, path planning, and collision avoidance in dynamic and complex environments.
Explores the development and simulation of coordinated drone operations using AirSim, an advanced physics-based platform. The project focuses on designing, testing, and optimizing multi-drone behaviors for tasks such as search-and-rescue, mapping, and environmental monitoring. Leveraging AirSim's simulation capabilities, the study aims to improve swarm intelligence, path planning, and collision avoidance in dynamic and complex environments.
Warfare Drone
Wisnu Jatmiko, Ario Yudo Husodo, Aprinaldi Jasa Mantau, et al.
Leverages the AirSim simulation tool to design and evaluate autonomous drones for modern battlefield scenarios. This project focuses on developing advanced strategies for reconnaissance, target tracking, and tactical engagement using AI-driven decision-making. By simulating combat environments, the study aims to refine algorithms for precision strikes, swarm coordination, and obstacle navigation under diverse terrain and adversarial conditions, contributing to the next generation of autonomous defense technologies.
Leverages the AirSim simulation tool to design and evaluate autonomous drones for modern battlefield scenarios. This project focuses on developing advanced strategies for reconnaissance, target tracking, and tactical engagement using AI-driven decision-making. By simulating combat environments, the study aims to refine algorithms for precision strikes, swarm coordination, and obstacle navigation under diverse terrain and adversarial conditions, contributing to the next generation of autonomous defense technologies.
Surveilance Object Detection
Wisnu Jatmiko, Grafika Jati, Aprinaldi Jasa Mantau, et al.
Focuses on developing advanced drone-based object detection systems for real-time surveillance applications. This project emphasizes the use of cutting-edge computer vision and machine learning techniques to identify, classify, and track objects in diverse environments. By utilizing real-world data and hardware testing, the research aims to enhance the precision and reliability of surveillance systems, catering to applications such as security monitoring, traffic analysis, and disaster management.
Focuses on developing advanced drone-based object detection systems for real-time surveillance applications. This project emphasizes the use of cutting-edge computer vision and machine learning techniques to identify, classify, and track objects in diverse environments. By utilizing real-world data and hardware testing, the research aims to enhance the precision and reliability of surveillance systems, catering to applications such as security monitoring, traffic analysis, and disaster management.
Reinforcement Learning
Wisnu Jatmiko, Vektor Dewanto, Aprinaldi Jasa Mantau, et al.
Investigates the development and application of reinforcement learning algorithms to solve complex decision-making problems across diverse domains. This project focuses on designing and optimizing RL models for tasks such as autonomous control, resource allocation, game strategies, and robotics. By leveraging advanced computational techniques, the research aims to enhance the scalability, adaptability, and efficiency of RL, contributing to breakthroughs in fields such as artificial intelligence, automation, and real-world problem-solving.
Investigates the development and application of reinforcement learning algorithms to solve complex decision-making problems across diverse domains. This project focuses on designing and optimizing RL models for tasks such as autonomous control, resource allocation, game strategies, and robotics. By leveraging advanced computational techniques, the research aims to enhance the scalability, adaptability, and efficiency of RL, contributing to breakthroughs in fields such as artificial intelligence, automation, and real-world problem-solving.
Renewable energy
Wisnu Jatmiko, Vektor Dewanto
Solar panel in every Indonesian rooftop, generation potential/prediction and challenges
Solar panel in every Indonesian rooftop, generation potential/prediction and challenges
IS for Biomedicine (IS4Biomed)
Heart failure detection
Wisnu Jatmiko, Muhammad Febrian Rachmadi
Focuses on developing advanced methods for early and accurate detection of heart failure using machine learning and data analytics. By analyzing medical data such as electrocardiograms (ECGs), echocardiograms, and patient health records, the project aims to identify critical patterns and risk factors associated with heart failure. The research emphasizes creating efficient, non-invasive, and reliable diagnostic tools to support clinicians in improving patient outcomes and enabling timely medical interventions. Applications include preventive healthcare, remote monitoring, and personalized treatment planning.
Focuses on developing advanced methods for early and accurate detection of heart failure using machine learning and data analytics. By analyzing medical data such as electrocardiograms (ECGs), echocardiograms, and patient health records, the project aims to identify critical patterns and risk factors associated with heart failure. The research emphasizes creating efficient, non-invasive, and reliable diagnostic tools to support clinicians in improving patient outcomes and enabling timely medical interventions. Applications include preventive healthcare, remote monitoring, and personalized treatment planning.
Left ventricle segmentation
Wisnu Jatmiko, Muhammad Febrian Rachmadi
Focuses on developing precise and automated methods for segmenting the left ventricle from medical imaging data, such as echocardiograms, CT, or MRI scans. This project employs advanced deep learning and image processing techniques to accurately delineate the structure of the left ventricle, enabling quantitative assessment of cardiac function. Applications include diagnosis and monitoring of heart diseases, treatment planning, and improving the accuracy of cardiac simulations. The research aims to enhance the efficiency and reliability of cardiac imaging analysis in clinical practice.
Focuses on developing precise and automated methods for segmenting the left ventricle from medical imaging data, such as echocardiograms, CT, or MRI scans. This project employs advanced deep learning and image processing techniques to accurately delineate the structure of the left ventricle, enabling quantitative assessment of cardiac function. Applications include diagnosis and monitoring of heart diseases, treatment planning, and improving the accuracy of cardiac simulations. The research aims to enhance the efficiency and reliability of cardiac imaging analysis in clinical practice.
Ejection fraction regression
Wisnu Jatmiko, Muhammad Febrian Rachmadi
Focuses on developing advanced machine learning models to estimate ejection fraction (EF) from medical imaging and clinical data. Ejection fraction, a key indicator of cardiac function, is crucial for diagnosing and monitoring heart conditions such as heart failure. This project leverages data-driven regression techniques to predict EF with high precision from echocardiograms, MRI scans, or other cardiac imaging modalities. The research aims to enhance the accuracy and efficiency of EF estimation, supporting clinicians in making informed decisions for patient care and treatment planning.
Focuses on developing advanced machine learning models to estimate ejection fraction (EF) from medical imaging and clinical data. Ejection fraction, a key indicator of cardiac function, is crucial for diagnosing and monitoring heart conditions such as heart failure. This project leverages data-driven regression techniques to predict EF with high precision from echocardiograms, MRI scans, or other cardiac imaging modalities. The research aims to enhance the accuracy and efficiency of EF estimation, supporting clinicians in making informed decisions for patient care and treatment planning.
Embryo grading
Wisnu Jatmiko, Muhammad Febrian Rachmadi, Aprinaldi Jasa Mantau, et al.
Focuses on developing advanced techniques for automated and objective assessment of embryo quality in assisted reproductive technologies, such as in vitro fertilization (IVF). This project employs state-of-the-art computer vision and machine learning algorithms to analyze microscopic images or time-lapse videos of embryos, evaluating morphological features and developmental patterns. The goal is to improve the accuracy, consistency, and efficiency of embryo grading, aiding embryologists in selecting the best candidates for implantation and enhancing the success rates of fertility treatments.
Focuses on developing advanced techniques for automated and objective assessment of embryo quality in assisted reproductive technologies, such as in vitro fertilization (IVF). This project employs state-of-the-art computer vision and machine learning algorithms to analyze microscopic images or time-lapse videos of embryos, evaluating morphological features and developmental patterns. The goal is to improve the accuracy, consistency, and efficiency of embryo grading, aiding embryologists in selecting the best candidates for implantation and enhancing the success rates of fertility treatments.
Brain white matter segmentation
Wisnu Jatmiko, Muhammad Febrian Rachmadi
Focuses on developing advanced methodologies for accurately segmenting white matter regions from brain imaging data, such as MRI scans. Utilizing cutting-edge deep learning and image processing techniques, the project aims to delineate white matter structures to support research in neuroanatomy, neurological disorders, and brain development. Applications include the study of diseases like multiple sclerosis, Alzheimer’s, and stroke, as well as aiding in surgical planning and cognitive neuroscience. The research seeks to enhance the precision and automation of white matter segmentation in clinical and research settings.
Focuses on developing advanced methodologies for accurately segmenting white matter regions from brain imaging data, such as MRI scans. Utilizing cutting-edge deep learning and image processing techniques, the project aims to delineate white matter structures to support research in neuroanatomy, neurological disorders, and brain development. Applications include the study of diseases like multiple sclerosis, Alzheimer’s, and stroke, as well as aiding in surgical planning and cognitive neuroscience. The research seeks to enhance the precision and automation of white matter segmentation in clinical and research settings.
Skin disease recognition
Wisnu Jatmiko, Muhammad Febrian Rachmadi
Focuses on developing advanced systems for accurately identifying and classifying various skin diseases using image-based analysis. Leveraging deep learning and computer vision techniques, the project processes dermoscopic images or photographs to detect conditions such as melanoma, eczema, psoriasis, and acne. The goal is to create a reliable, non-invasive, and efficient diagnostic tool that supports dermatologists and primary care providers, improves early detection, and enhances patient outcomes. Applications include telemedicine, remote diagnostics, and personalized treatment recommendations.
Focuses on developing advanced systems for accurately identifying and classifying various skin diseases using image-based analysis. Leveraging deep learning and computer vision techniques, the project processes dermoscopic images or photographs to detect conditions such as melanoma, eczema, psoriasis, and acne. The goal is to create a reliable, non-invasive, and efficient diagnostic tool that supports dermatologists and primary care providers, improves early detection, and enhances patient outcomes. Applications include telemedicine, remote diagnostics, and personalized treatment recommendations.
Liver segmentation
Wisnu Jatmiko, Muhammad Febrian Rachmadi
Focuses on developing precise and automated techniques for segmenting the liver from medical imaging modalities such as CT and MRI scans. Utilizing advanced deep learning and image processing algorithms, the project aims to accurately delineate liver boundaries to support applications such as liver disease diagnosis, surgical planning, and treatment monitoring. By enhancing the accuracy and efficiency of liver segmentation, the research contributes to improved clinical workflows and better outcomes in hepatology and radiology practices.
Focuses on developing precise and automated techniques for segmenting the liver from medical imaging modalities such as CT and MRI scans. Utilizing advanced deep learning and image processing algorithms, the project aims to accurately delineate liver boundaries to support applications such as liver disease diagnosis, surgical planning, and treatment monitoring. By enhancing the accuracy and efficiency of liver segmentation, the research contributes to improved clinical workflows and better outcomes in hepatology and radiology practices.
Triase decision making
Wisnu Jatmiko, Muhammad Febrian Rachmadi
Focuses on developing intelligent systems to support rapid and accurate triage in emergency and clinical settings. This project leverages machine learning, decision support algorithms, and real-time patient data analysis to prioritize care based on the severity of a patient’s condition. The goal is to improve the efficiency and consistency of triage processes, ensuring timely interventions and optimal resource allocation. Applications include emergency departments, disaster response scenarios, and telemedicine, ultimately enhancing patient outcomes and healthcare delivery.
Focuses on developing intelligent systems to support rapid and accurate triage in emergency and clinical settings. This project leverages machine learning, decision support algorithms, and real-time patient data analysis to prioritize care based on the severity of a patient’s condition. The goal is to improve the efficiency and consistency of triage processes, ensuring timely interventions and optimal resource allocation. Applications include emergency departments, disaster response scenarios, and telemedicine, ultimately enhancing patient outcomes and healthcare delivery.
Continual Learning For Biomedical Image Analysis
Wisnu Jatmiko, Muhammad Febrian Rachmadi
Focuses on developing advanced machine learning methods to enable models to learn and adapt incrementally from new biomedical imaging data without forgetting previously acquired knowledge. This approach addresses challenges such as domain shifts, dataset variability, and evolving diagnostic needs in biomedical imaging modalities like MRI, CT, and histopathology. The project aims to create robust and scalable systems that enhance diagnostic accuracy and efficiency across diverse healthcare applications, supporting real-time adaptation in clinical and research environments.
Focuses on developing advanced machine learning methods to enable models to learn and adapt incrementally from new biomedical imaging data without forgetting previously acquired knowledge. This approach addresses challenges such as domain shifts, dataset variability, and evolving diagnostic needs in biomedical imaging modalities like MRI, CT, and histopathology. The project aims to create robust and scalable systems that enhance diagnostic accuracy and efficiency across diverse healthcare applications, supporting real-time adaptation in clinical and research environments.
IS for Surveilance/Remote Sensing (IS4SRS)
Building damage assessment
Wisnu Jatmiko, Aprinaldi Jasa Mantau, Mgs M Luthfi Ramadhan, et al.
Focuses on developing advanced methods to evaluate structural damage in buildings following disasters such as earthquakes, floods, or fires. This project utilizes state-of-the-art computer vision, machine learning, and data analytics techniques to analyze images, videos, and sensor data for detecting and classifying damage. The goal is to create efficient, automated tools for rapid and accurate damage assessment, aiding in emergency response, recovery planning, and improving disaster resilience in urban and rural infrastructures.
Focuses on developing advanced methods to evaluate structural damage in buildings following disasters such as earthquakes, floods, or fires. This project utilizes state-of-the-art computer vision, machine learning, and data analytics techniques to analyze images, videos, and sensor data for detecting and classifying damage. The goal is to create efficient, automated tools for rapid and accurate damage assessment, aiding in emergency response, recovery planning, and improving disaster resilience in urban and rural infrastructures.
Change detection
Wisnu Jatmiko, Aprinaldi Jasa Mantau, Yohannes F. Hestrio, et al.
Developing advanced techniques to identify and analyze changes in environments over time using image and data analysis. This project employs cutting-edge machine learning and remote sensing technologies to detect changes such as land use alterations, urban development, deforestation, and disaster impacts. By processing and comparing multi-temporal datasets, the research aims to create robust tools for applications in environmental monitoring, infrastructure management, and urban planning, enabling proactive decision-making and resource management.
Developing advanced techniques to identify and analyze changes in environments over time using image and data analysis. This project employs cutting-edge machine learning and remote sensing technologies to detect changes such as land use alterations, urban development, deforestation, and disaster impacts. By processing and comparing multi-temporal datasets, the research aims to create robust tools for applications in environmental monitoring, infrastructure management, and urban planning, enabling proactive decision-making and resource management.
Agricultural segmentation
Wisnu Jatmiko, Aprinaldi Jasa Mantau, Yohannes F. Hestrio, et al.
Focuses on developing precise methods for segmenting and analyzing agricultural land using advanced image processing and machine learning techniques. This project aims to classify and map different land-use types, crop patterns, and field boundaries from satellite or aerial imagery. By improving the accuracy and efficiency of agricultural land segmentation, the research supports better resource allocation, crop monitoring, and sustainable farming practices, ultimately contributing to enhanced agricultural productivity and environmental management.
Focuses on developing precise methods for segmenting and analyzing agricultural land using advanced image processing and machine learning techniques. This project aims to classify and map different land-use types, crop patterns, and field boundaries from satellite or aerial imagery. By improving the accuracy and efficiency of agricultural land segmentation, the research supports better resource allocation, crop monitoring, and sustainable farming practices, ultimately contributing to enhanced agricultural productivity and environmental management.
Building detection
Wisnu Jatmiko, Aprinaldi Jasa Mantau, Yohannes F. Hestrio, et al.
Focuses on developing advanced algorithms to identify and map buildings from aerial or satellite imagery with high accuracy. Utilizing techniques in computer vision and machine learning, this project aims to automate the detection and classification of structures for applications such as urban planning, disaster response, and infrastructure monitoring. By enhancing the precision and scalability of building detection methods, the research contributes to smarter city management, efficient resource allocation, and improved disaster resilience strategies.
Focuses on developing advanced algorithms to identify and map buildings from aerial or satellite imagery with high accuracy. Utilizing techniques in computer vision and machine learning, this project aims to automate the detection and classification of structures for applications such as urban planning, disaster response, and infrastructure monitoring. By enhancing the precision and scalability of building detection methods, the research contributes to smarter city management, efficient resource allocation, and improved disaster resilience strategies.
LIDAR point cloud
Wisnu Jatmiko, Aprinaldi Jasa Mantau, Mgs M Luthfi Ramadhan, et al.
Focuses on leveraging LiDAR technology to analyze and interpret high-resolution 3D point cloud data for various applications. This project explores techniques for object detection, segmentation, and classification within LiDAR datasets to enable precise modeling of terrains, structures, and vegetation. Applications include urban mapping, autonomous navigation, disaster assessment, and environmental monitoring. By advancing processing and analysis methodologies, the research aims to enhance the usability and accuracy of LiDAR data across industries.
Focuses on leveraging LiDAR technology to analyze and interpret high-resolution 3D point cloud data for various applications. This project explores techniques for object detection, segmentation, and classification within LiDAR datasets to enable precise modeling of terrains, structures, and vegetation. Applications include urban mapping, autonomous navigation, disaster assessment, and environmental monitoring. By advancing processing and analysis methodologies, the research aims to enhance the usability and accuracy of LiDAR data across industries.
IS for Life and Social Sciences (IS4LSS)
Emotion Recognition Based on 3D Generated Image
Wisnu Jatmiko, Aprinaldi Jasa Mantau, Hannan Hunafa, et al.
Focuses on developing a system to accurately analyze and identify human emotions using 3D facial imagery. This project utilizes three cameras to capture multiple perspectives, enabling the creation of detailed 3D facial models. The data is processed on a Jetson Nano edge device, leveraging its computational efficiency for real-time analysis. By integrating computer vision and deep learning algorithms, the system detects and interprets subtle facial expressions with high precision. Applications include human-computer interaction, mental health assessment, and virtual reality, with a focus on achieving robust performance in real-world scenarios.
Focuses on developing a system to accurately analyze and identify human emotions using 3D facial imagery. This project utilizes three cameras to capture multiple perspectives, enabling the creation of detailed 3D facial models. The data is processed on a Jetson Nano edge device, leveraging its computational efficiency for real-time analysis. By integrating computer vision and deep learning algorithms, the system detects and interprets subtle facial expressions with high precision. Applications include human-computer interaction, mental health assessment, and virtual reality, with a focus on achieving robust performance in real-world scenarios.
AI tutor for every Indonesian kid
Wisnu Jatmiko, Vektor Dewanto
Developing AI for personalized learning
Developing AI for personalized learning