David K. Han
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
David K. Han has a total of 48 co-authors. Here are the top co-authors by number of shared publications:
- Hanseok Ko
Korea University
17 shared publications - Amirreza Rouhi
Drexel University
3 shared publications - Bonhwa Ku
School of Electrical Engineering, Korea University, Seoul, South Korea
3 shared publications - Youngsaeng Jin3 shared publications
- Jaihyun Park
School of Electrical Engineering, Korea University, Anam-dong, Seongbuk-gu, Seoul, 136-713, Korea
2 shared publications - Jeong-Gi Kwak
Korea Univerity Seoul Korea
2 shared publications
Latest Publications
- 30 June 2026
- Published byĀ Drexel University Libraries
Robust visual perception in challenging real-world scenarios
This dissertation investigates robust visual perception for detecting and recognizing difficult objects in challenging real-world environments, tracing a single arc from long-range small-UAV detection to open-ended object detection. The work began with the practical problem of detecting small flying objects at long range, where drones routinely occupy fewer than fifty pixels, blend into cluttered backgrounds, and span distances from a few metres to over a hundred. The deeper obstacle was data: no public drone-detection benchmark captured the operating conditions we needed, and none provided camera-to-drone range information. We therefore contributed a comprehensive survey of deep learning methods for drone detection, covering 16 datasets and 24 algorithms, and introduced two new benchmarks: the Long-Range Drone Detection Dataset (LRDD) with 22,500 images, and its enhanced successor LRDDv2 with 39,516 images and explicit camera-to-drone range annotations for over 8,000 images, the first such information in any public drone-detection dataset. On standard benchmarks, models trained on LRDD and LRDDv2 substantially outperform baselines trained on Drone-vs-Bird, with the largest gains in the small-bounding-box regime that defines long-range detection. Yet the long-range drone problem surfaced a deeper limit that better datasets alone cannot lift. When an object's visual signature shrinks to a few dozen pixels and shares texture, motion, or shape with its surroundings, the information needed to identify it is not fully present in the image. And drone detection turns out to be a relatively constrained instance of this bottleneck: at altitude the realistic candidate vocabulary is small, including drones, birds, and perhaps a handful of airborne categories, so even when pixels degrade, the space of plausible answers stays tightly bounded. General object detection enjoys no such luxury. A street scene at dusk admits dozens of plausible categories, an indoor scene hundreds, and an open-ended setting no a-priori bound at all. Under fog, motion blur, partial occlusion, or low light, the same pixel degradation that hurts drone detection compounds in the general case, weakening evidence for the right answer while multiplying the alternatives that remain consistent with it. Humans cope with these regimes by drawing on knowledge that lies outside the frame: the spatial layout of the scene, the categories that typically co-occur there, and the implicit physics of what can plausibly appear and where. The remainder of this dissertation pursues that observation, asking how vision systems can incorporate the kind of scene-level contextual knowledge that humans take for granted: first to rescue detections of familiar categories under adverse conditions, and ultimately to recognise objects from categories never encountered at training time. We first present LMOD, a transformer-based model that learns scene-wide contextual relationships exclusively from object labels, positions, and sizes, with no pixel access at all. Integrating LMOD with conventional detectors such as DETR and YOLOv8 yields consistent improvements on COCO, Visual Genome, and the challenging ExDark low-light dataset, with the largest gains under degraded imaging conditions. We then scale the contextual paradigm by coupling YOLO detectors with LLaMA 2, a Large Language Model, which validates low-confidence detections through scene-level contextual prompts. Experiments on COCO-2017 and the newly introduced COCO-2017-Blurred benchmark confirm significant gains, particularly for small and occluded objects. We then extend the work toward open-ended object detection, the ability to both detect and assign meaningful semantic labels to previously unseen objects without any predefined vocabulary. ADAM establishes the core architecture by constructing an Embedding-Label Repository (ELR) that pairs CLIP visual embeddings of unknown objects with candidate labels generated through context-aware LLM prompting; on COCO validation, ADAM reaches 61.3% Top-1 accuracy and transfers training-free to PASCAL VOC. COSRA extends ADAM with a complete detection front-end, using SAM and Faster R-CNN, attribute-based characterization across 11 categories and 155 descriptors, and an iterative semantic-grouping refinement mechanism. On COCO, it achieves 38.8% Novel AP, and on LVIS v1, it achieves 24.2% Rare AP, surpassing every open-vocabulary baseline despite requiring neither training nor a vocabulary at inference. Our most advanced framework upgrades the detection front-end to Mask R-CNN and introduces diffusion-generated confusion anchors: for each unknown object, Stable Diffusion XL synthesizes contrastive visual exemplars from second-ranked LLM predictions, localized by Florence-2 and validated by CLIP, enriching the ELR with the representational diversity needed to disambiguate visually similar categories. A cross-label iterative refinement mechanism then groups semantically similar labels and corrects errors across related categories through k-NN majority voting. On COCO and LVIS minival, this framework achieves 60.8% Novel AP50 and 52.4% APr respectively, surpassing all existing open-ended and open-vocabulary methods, including those with larger backbones and extensive supervised training, while requiring neither retraining nor vocabulary supervision. Collectively, this body of work establishes a unified framework for robust visual perception of small, rare, and novel objects in complex real-world environments, progressing from specialised long-range drone detection through context-aware closed-set detection to fully open-ended scene understanding.
- 30 June 2026
- Published byĀ Drexel University Libraries
Robust visual perception in challenging real-world scenarios
This dissertation investigates robust visual perception for detecting and recognizing difficult objects in challenging real-world environments, tracing a single arc from long-range small-UAV detection to open-ended object detection. The work began with the practical problem of detecting small flying objects at long range, where drones routinely occupy fewer than fifty pixels, blend into cluttered backgrounds, and span distances from a few metres to over a hundred. The deeper obstacle was data: no public drone-detection benchmark captured the operating conditions we needed, and none provided camera-to-drone range information. We therefore contributed a comprehensive survey of deep learning methods for drone detection, covering 16 datasets and 24 algorithms, and introduced two new benchmarks: the Long-Range Drone Detection Dataset (LRDD) with 22,500 images, and its enhanced successor LRDDv2 with 39,516 images and explicit camera-to-drone range annotations for over 8,000 images, the first such information in any public drone-detection dataset. On standard benchmarks, models trained on LRDD and LRDDv2 substantially outperform baselines trained on Drone-vs-Bird, with the largest gains in the small-bounding-box regime that defines long-range detection. Yet the long-range drone problem surfaced a deeper limit that better datasets alone cannot lift. When an object's visual signature shrinks to a few dozen pixels and shares texture, motion, or shape with its surroundings, the information needed to identify it is not fully present in the image. And drone detection turns out to be a relatively constrained instance of this bottleneck: at altitude the realistic candidate vocabulary is small, including drones, birds, and perhaps a handful of airborne categories, so even when pixels degrade, the space of plausible answers stays tightly bounded. General object detection enjoys no such luxury. A street scene at dusk admits dozens of plausible categories, an indoor scene hundreds, and an open-ended setting no a-priori bound at all. Under fog, motion blur, partial occlusion, or low light, the same pixel degradation that hurts drone detection compounds in the general case, weakening evidence for the right answer while multiplying the alternatives that remain consistent with it. Humans cope with these regimes by drawing on knowledge that lies outside the frame: the spatial layout of the scene, the categories that typically co-occur there, and the implicit physics of what can plausibly appear and where. The remainder of this dissertation pursues that observation, asking how vision systems can incorporate the kind of scene-level contextual knowledge that humans take for granted: first to rescue detections of familiar categories under adverse conditions, and ultimately to recognise objects from categories never encountered at training time. We first present LMOD, a transformer-based model that learns scene-wide contextual relationships exclusively from object labels, positions, and sizes, with no pixel access at all. Integrating LMOD with conventional detectors such as DETR and YOLOv8 yields consistent improvements on COCO, Visual Genome, and the challenging ExDark low-light dataset, with the largest gains under degraded imaging conditions. We then scale the contextual paradigm by coupling YOLO detectors with LLaMA 2, a Large Language Model, which validates low-confidence detections through scene-level contextual prompts. Experiments on COCO-2017 and the newly introduced COCO-2017-Blurred benchmark confirm significant gains, particularly for small and occluded objects. We then extend the work toward open-ended object detection, the ability to both detect and assign meaningful semantic labels to previously unseen objects without any predefined vocabulary. ADAM establishes the core architecture by constructing an Embedding-Label Repository (ELR) that pairs CLIP visual embeddings of unknown objects with candidate labels generated through context-aware LLM prompting; on COCO validation, ADAM reaches 61.3% Top-1 accuracy and transfers training-free to PASCAL VOC. COSRA extends ADAM with a complete detection front-end, using SAM and Faster R-CNN, attribute-based characterization across 11 categories and 155 descriptors, and an iterative semantic-grouping refinement mechanism. On COCO, it achieves 38.8% Novel AP, and on LVIS v1, it achieves 24.2% Rare AP, surpassing every open-vocabulary baseline despite requiring neither training nor a vocabulary at inference. Our most advanced framework upgrades the detection front-end to Mask R-CNN and introduces diffusion-generated confusion anchors: for each unknown object, Stable Diffusion XL synthesizes contrastive visual exemplars from second-ranked LLM predictions, localized by Florence-2 and validated by CLIP, enriching the ELR with the representational diversity needed to disambiguate visually similar categories. A cross-label iterative refinement mechanism then groups semantically similar labels and corrects errors across related categories through k-NN majority voting. On COCO and LVIS minival, this framework achieves 60.8% Novel AP50 and 52.4% APr respectively, surpassing all existing open-ended and open-vocabulary methods, including those with larger backbones and extensive supervised training, while requiring neither retraining nor vocabulary supervision. Collectively, this body of work establishes a unified framework for robust visual perception of small, rare, and novel objects in complex real-world environments, progressing from specialised long-range drone detection through context-aware closed-set detection to fully open-ended scene understanding.
- Open Access
Resnet-conformer network with shared weights and attention mechanism for sound event localization, detection, and distance estimation
This technical report outlines our approach to Task 3A of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2024, focusing on Sound Event Localization and Detection (SELD). SELD provides valuable insights by estimating sound event localization and detection, aiding in various machine cognition tasks such as environmental inference, navigation, and other sound localization-related applications. This year's challenge evaluates models using either audio-only (Track A) or audiovisual (Track B) inputs on annotated recordings of real sound scenes. A notable change this year is the introduction of distance estimation, with evaluation metrics adjusted accordingly for a comprehensive assessment. Our submission is for Task A of the Challenge, which focuses on the audio-only track. Our approach utilizes log-mel spectrograms, intensity vectors, and employs multiple data augmentations. We proposed an EINV2-based [1] network architecture, achieving improved results: an F-score of 40.2%, Angular Error (DOA) of 17.7 degrees, and Relative Distance Error (RDE) of 0.32 on the test set of the Development Dataset [2 ,3].
Open AccessResnet-conformer network with shared weights and attention mechanism for sound event localization, detection, and distance estimation
This technical report outlines our approach to Task 3A of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2024, focusing on Sound Event Localization and Detection (SELD). SELD provides valuable insights by estimating sound event localization and detection, aiding in various machine cognition tasks such as environmental inference, navigation, and other sound localization-related applications. This year's challenge evaluates models using either audio-only (Track A) or audiovisual (Track B) inputs on annotated recordings of real sound scenes. A notable change this year is the introduction of distance estimation, with evaluation metrics adjusted accordingly for a comprehensive assessment. Our submission is for Task A of the Challenge, which focuses on the audio-only track. Our approach utilizes log-mel spectrograms, intensity vectors, and employs multiple data augmentations. We proposed an EINV2-based [1] network architecture, achieving improved results: an F-score of 40.2%, Angular Error (DOA) of 17.7 degrees, and Relative Distance Error (RDE) of 0.32 on the test set of the Development Dataset [2 ,3].
- 3 December 2024
- Published byĀ Springer Nature
- p. 299-314
Enhancing Object Detection by Leveraging Large Language Models for Contextual Knowledge
- 3 December 2024
- Published byĀ Springer Nature
- p. 299-314
Enhancing Object Detection by Leveraging Large Language Models for Contextual Knowledge
- 3 December 2024
- Published byĀ Springer Nature
- p. 366-380
Data Augmentation Pipeline for Enhanced UAV Surveillance
- 3 December 2024
- Published byĀ Springer Nature
- p. 366-380
Data Augmentation Pipeline for Enhanced UAV Surveillance
- 3 March 2024
- Published byĀ ElsevierĀ inĀ Pattern Recognition Letters
- Vol.Ā 180, 68-74
Towards high-fidelity facial UV map generation in real-world
- 3 March 2024
- Published byĀ ElsevierĀ inĀ Pattern Recognition Letters
- Vol.Ā 180, 68-74
Towards high-fidelity facial UV map generation in real-world
- 25 January 2024
- Published byĀ Drexel University Libraries
Dataset augmentation pipeline for improved long-range drone detection
This thesis addresses the critical need for accurate drone detection models in response to the widespread use of Unmanned Aerial Vehicles (UAVs) across various industries which introduces safety risks and security concerns, highlighting the importance of developing robust drone detection algorithms. To address these challenges, this thesis presents a two-phase solution for long-range drone detection. The first phase focuses on seamlessly integrating synthetic data, generated with Unreal Engine, with real-world data to enhance dataset diversity. In the subsequent phase, the Cycle-consistent Generative Adversarial Network (CycleGAN) model is employed to translate synthetic images into the real domain, effectively bridging the gap between synthetic and real data. This comprehensive approach aims to advance drone detection models in terms of precision while also providing a cost-effective and accessible method for generating diverse drone detection datasets. To assess the effectiveness of the proposed pipeline for data augmentation, the You Only Look Once (YOLO) model is trained in three scenarios: using only real data, augmenting real data with synthetic data, and augmenting real data with translated data. The trained models are then tested on three unseen real datasets: DetFly, Drone Detection, and UAV Detect. The results conclusively demonstrate that including both synthetic and translated images in the real dataset enhances drone detection capabilities. Notably, the most substantial improvement is observed when augmenting real data with translated data. These findings emphasize the significance of the proposed approach in enhancing drone detection performance.
- 25 January 2024
- Published byĀ Drexel University Libraries
Dataset augmentation pipeline for improved long-range drone detection
This thesis addresses the critical need for accurate drone detection models in response to the widespread use of Unmanned Aerial Vehicles (UAVs) across various industries which introduces safety risks and security concerns, highlighting the importance of developing robust drone detection algorithms. To address these challenges, this thesis presents a two-phase solution for long-range drone detection. The first phase focuses on seamlessly integrating synthetic data, generated with Unreal Engine, with real-world data to enhance dataset diversity. In the subsequent phase, the Cycle-consistent Generative Adversarial Network (CycleGAN) model is employed to translate synthetic images into the real domain, effectively bridging the gap between synthetic and real data. This comprehensive approach aims to advance drone detection models in terms of precision while also providing a cost-effective and accessible method for generating diverse drone detection datasets. To assess the effectiveness of the proposed pipeline for data augmentation, the You Only Look Once (YOLO) model is trained in three scenarios: using only real data, augmenting real data with synthetic data, and augmenting real data with translated data. The trained models are then tested on three unseen real datasets: DetFly, Drone Detection, and UAV Detect. The results conclusively demonstrate that including both synthetic and translated images in the real dataset enhances drone detection capabilities. Notably, the most substantial improvement is observed when augmenting real data with translated data. These findings emphasize the significance of the proposed approach in enhancing drone detection performance.
- Open Access
Learning Scene Context Without Images
Teaching machines of scene contextual knowledge would enable them to interact more effectively with the environment and to anticipate or predict objects that may not be immediately apparent in their perceptual field. In this paper, we introduce a novel transformer-based approach called $LMOD$ ( Label-based Missing Object Detection) to teach scene contextual knowledge to machines using an attention mechanism. A distinctive aspect of the proposed approach is its reliance solely on labels from image datasets to teach scene context, entirely eliminating the need for the actual image itself. We show how scene-wide relationships among different objects can be learned using a self-attention mechanism. We further show that the contextual knowledge gained from label based learning can enhance performance of other visual based object detection algorithm.Open AccessLearning Scene Context Without Images
Teaching machines of scene contextual knowledge would enable them to interact more effectively with the environment and to anticipate or predict objects that may not be immediately apparent in their perceptual field. In this paper, we introduce a novel transformer-based approach called $LMOD$ ( Label-based Missing Object Detection) to teach scene contextual knowledge to machines using an attention mechanism. A distinctive aspect of the proposed approach is its reliance solely on labels from image datasets to teach scene context, entirely eliminating the need for the actual image itself. We show how scene-wide relationships among different objects can be learned using a self-attention mechanism. We further show that the contextual knowledge gained from label based learning can enhance performance of other visual based object detection algorithm. - 3 October 2023
- Published byĀ Drexel University Libraries
Acoustic Signal Classification in Underwater and Air Environments
Identifying various acoustic sources can be challenging in both air and underwater settings. This is largely due to the difficulties in data collection, especially in ocean environments, and the variety of background noise they pose. Transformer architecture with self-attention mechanism has proven effective across various machine learning tasks, such as Natural Language Processing (NLP) and image recognition, even in challenging environments. This thesis presents the applications of Transformer in both air and underwater domains. In the air, we developed a Hierarchical Knowledge Distillation (H-KD) framework that utilized the Audio Spectrogram Transformer (AST) [1] architecture as a teacher model. The framework was designed using a multi-stage learning approach and cross-model integration to enhance performance of a low complexity Convolutional Neural Networks (CNN) student model. We applied it to an air acoustic dataset - TAU Urban Acoustic Scenes 2023 Mobile development [2] and demonstrated that it achieved a significantly improved performance compared to the Baseline system [3]. After verifying the model on publicly available datasets, our proposed model was submitted to the 2023 IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE2023) [3]. In the underwater acoustic signals experiment, we modified and adapted the Hierarchical Token Semantic - Audio Transformer (HTS-AT) [4] architecture to make it better suited underwater environments. We applied it to an underwater acoustic dataset - shipsEar [5] and demonstrated that our proposed method outperformed some of the latest classification methods. The promising results obtained motivate further exploration of Transformer-based approaches in future research to advance the state-of-the-art (SOTA) in Acoustic Signal Classifications (ASC).
- 3 October 2023
- Published byĀ Drexel University Libraries
Acoustic Signal Classification in Underwater and Air Environments
Identifying various acoustic sources can be challenging in both air and underwater settings. This is largely due to the difficulties in data collection, especially in ocean environments, and the variety of background noise they pose. Transformer architecture with self-attention mechanism has proven effective across various machine learning tasks, such as Natural Language Processing (NLP) and image recognition, even in challenging environments. This thesis presents the applications of Transformer in both air and underwater domains. In the air, we developed a Hierarchical Knowledge Distillation (H-KD) framework that utilized the Audio Spectrogram Transformer (AST) [1] architecture as a teacher model. The framework was designed using a multi-stage learning approach and cross-model integration to enhance performance of a low complexity Convolutional Neural Networks (CNN) student model. We applied it to an air acoustic dataset - TAU Urban Acoustic Scenes 2023 Mobile development [2] and demonstrated that it achieved a significantly improved performance compared to the Baseline system [3]. After verifying the model on publicly available datasets, our proposed model was submitted to the 2023 IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE2023) [3]. In the underwater acoustic signals experiment, we modified and adapted the Hierarchical Token Semantic - Audio Transformer (HTS-AT) [4] architecture to make it better suited underwater environments. We applied it to an underwater acoustic dataset - shipsEar [5] and demonstrated that our proposed method outperformed some of the latest classification methods. The promising results obtained motivate further exploration of Transformer-based approaches in future research to advance the state-of-the-art (SOTA) in Acoustic Signal Classifications (ASC).
- 3 October 2023
- Published byĀ Drexel University Libraries
Underwater Acoustic Target Recognition on ShipsEar Dataset
The task of classifying underwater audio source has various marine-oriented applications, including maritime and environmental monitoring, detection of marine life, and underwater surveillance. However, Underwater Acoustic Target Recognition (UATR) remains challenging due to several factors. These include high cost of acquiring acoustic data, lack of of publicly accessible labeled datasets, interference induced by both underwater and over-the-air noise sources, and other environmental factors such as sound speed variations, weather, bottom boundary conditions, etc. In this thesis, we propose a convolution-free, purely attention-based transformer architecture for the task of classifying underwater acoustic source. The architecture, adopted from the Audio Spectrogram Transformer (AST) applied so far for air acoustic setting, is capable of capturing long-range global context for enhanced underwater acoustic signal classification. Additionally, we further improve the performance by leveraging transfer learning from Vision Transformer (ViT), which has been pretrained on the ImageNet dataset. The AST model is trained on the ShipsEar dataset, which consists of 90 instances of ship-emitted underwater sounds including 11 types of vessels and one background noise category. These sound recordings, obtained under real conditions in shallow waters, encapsulate both of natural and anthropogenic. Currently, the state-of-the-art Convolutional Long Short-term Memory (ConvLSTM) applied to this task achieves an accuracy of 0.9878 and an F1 score of 0.9878. However, our proposed AST model surpasses the ConvLSTM performance, reaching an accuracy of 0.9890 and an F1 score of 0.9897
- 3 October 2023
- Published byĀ Drexel University Libraries
Underwater Acoustic Target Recognition on ShipsEar Dataset
The task of classifying underwater audio source has various marine-oriented applications, including maritime and environmental monitoring, detection of marine life, and underwater surveillance. However, Underwater Acoustic Target Recognition (UATR) remains challenging due to several factors. These include high cost of acquiring acoustic data, lack of of publicly accessible labeled datasets, interference induced by both underwater and over-the-air noise sources, and other environmental factors such as sound speed variations, weather, bottom boundary conditions, etc. In this thesis, we propose a convolution-free, purely attention-based transformer architecture for the task of classifying underwater acoustic source. The architecture, adopted from the Audio Spectrogram Transformer (AST) applied so far for air acoustic setting, is capable of capturing long-range global context for enhanced underwater acoustic signal classification. Additionally, we further improve the performance by leveraging transfer learning from Vision Transformer (ViT), which has been pretrained on the ImageNet dataset. The AST model is trained on the ShipsEar dataset, which consists of 90 instances of ship-emitted underwater sounds including 11 types of vessels and one background noise category. These sound recordings, obtained under real conditions in shallow waters, encapsulate both of natural and anthropogenic. Currently, the state-of-the-art Convolutional Long Short-term Memory (ConvLSTM) applied to this task achieves an accuracy of 0.9878 and an F1 score of 0.9878. However, our proposed AST model surpasses the ConvLSTM performance, reaching an accuracy of 0.9890 and an F1 score of 0.9897
- 3 October 2023
- Published byĀ Drexel University Libraries
Design and Fabrication of an Augmented Reality-Enabled, AI-Assisted Autonomous Mobile Robot with Dual Six-Axis Robotic Arms for Advanced Object Manipulation
This thesis presents the design and fabrication of an autonomous mobile robot with dual six-axis robotic arms, enhanced by augmented reality (AR) and powered by artificial intelligence (AI). The innovation aims to advance object manipulation tasks in various complex and dynamic environments, potentially for industrial applications. The system's Artificial Intelligence feature enhances the robot's ability to navigate and interact with its surroundings more accurately. The AR interface also allows human operators to intuitively control and guide the robot's actions in real time when needed, bridging the gap between autonomous operation and human oversight. The dual six-axis robotic arms enable dexterous and precise object manipulation. These arms mimic the versatility and adaptability of human arms, opening new possibilities for tasks that were previously challenging for traditional robotic systems. In developing this advanced mobile robot, this work embodies a pioneering step in autonomous robotics, showcasing a blend of AI, AR, and mechanical design. The robot's capabilities in object manipulation present a leap forward in robotics, promising a future where robots can take on more complex roles in our societies, while also offering a practical solution to labor-intensive or high-risk tasks. The research concludes with extensive testing and validation of the robot, demonstrating its effectiveness and efficiency in performing complex object manipulation tasks autonomously in various controlled and real-world scenarios. The results suggest promising implications for the future of robotics and automation, reinforcing the importance of interdisciplinary approaches in technological advancement.
- 3 October 2023
- Published byĀ Drexel University Libraries
Design and Fabrication of an Augmented Reality-Enabled, AI-Assisted Autonomous Mobile Robot with Dual Six-Axis Robotic Arms for Advanced Object Manipulation
This thesis presents the design and fabrication of an autonomous mobile robot with dual six-axis robotic arms, enhanced by augmented reality (AR) and powered by artificial intelligence (AI). The innovation aims to advance object manipulation tasks in various complex and dynamic environments, potentially for industrial applications. The system's Artificial Intelligence feature enhances the robot's ability to navigate and interact with its surroundings more accurately. The AR interface also allows human operators to intuitively control and guide the robot's actions in real time when needed, bridging the gap between autonomous operation and human oversight. The dual six-axis robotic arms enable dexterous and precise object manipulation. These arms mimic the versatility and adaptability of human arms, opening new possibilities for tasks that were previously challenging for traditional robotic systems. In developing this advanced mobile robot, this work embodies a pioneering step in autonomous robotics, showcasing a blend of AI, AR, and mechanical design. The robot's capabilities in object manipulation present a leap forward in robotics, promising a future where robots can take on more complex roles in our societies, while also offering a practical solution to labor-intensive or high-risk tasks. The research concludes with extensive testing and validation of the robot, demonstrating its effectiveness and efficiency in performing complex object manipulation tasks autonomously in various controlled and real-world scenarios. The results suggest promising implications for the future of robotics and automation, reinforcing the importance of interdisciplinary approaches in technological advancement.