Abhishek Phadke
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Abhishek Phadke has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
F. Antonio Medrano
Conrad Blucher Institute for Surveying and Science, Texas A&M University-Corpus Christi, Corpus Christi, TX, United States
10 shared publications- Fernando Antonio Medrano
Department of Computer Science Texas A&M University‐Corpus Christi Corpus Christi Texas USA
4 shared publications Stanislav Ustymenko
4 shared publications- Tianxing Chu
Conrad Blucher Institute for Surveying and Science, Texas A&M University-Corpus Christi, Corpus Christi, TX 78412, USA
4 shared publications - Chandra N. Sekharan
Department of Computer Science, Texas A&M University-Corpus Christi, Corpus Christi, TX 78412, USA
3 shared publications - Jahnavi Krishna Koda
Department of Coastal and Marine System Science Program, Texas A&M University-Corpus Christi, Corpus Christi, TX 78412, USA
3 shared publications
Latest Publications
- Open Access
- 13 October 2026
- Published by MDPI AG in Preprints.org
A Five-Level Operational Maturity Framework for Machine Learning in Midstream Oil and Gas
Machine learning is increasingly applied to midstream oil and gas pipelines, yet reported performance rarely establishes readiness for field use. This narrative review organizes peer-reviewed and industry sources from 2015–2025 using a five-level operational maturity framework (sensing and detection, diagnosis, prediction, decision and optimization, and human-supervised closed-loop automation), defined by a rubric of decision authority, operator role, and the validation evidence each level requires. A second axis records where each study’s performance was demonstrated, from simulation to closed-loop use. Applications include leak and anomaly detection, defect and corrosion assessment, hard spots, pump cavitation, predictive maintenance, failure-risk modeling, scheduling, and digital twins. Cross-classifying the coded sources on both axes shows that validation realism does not keep pace with claimed operational responsibility. Peer-reviewed evaluations are concentrated at Levels 1–3 and rely largely on laboratory, simulated, or retrospective data; the only operational Level 4 example is a practitioner case, and no coded source demonstrates Level 5 closed-loop control. The corpus thus shows a validation gap at the lower levels and an evidence-coverage gap at the upper ones; beyond the performance shown in controlled evaluations, validation, transferability, and governance remain major unresolved barriers to deployment. A ranked agenda led by graduated field validation follows.
- 13 October 2026
- Published by MDPI AG in Preprints.org
Open AccessA Five-Level Operational Maturity Framework for Machine Learning in Midstream Oil and Gas
Machine learning is increasingly applied to midstream oil and gas pipelines, yet reported performance rarely establishes readiness for field use. This narrative review organizes peer-reviewed and industry sources from 2015–2025 using a five-level operational maturity framework (sensing and detection, diagnosis, prediction, decision and optimization, and human-supervised closed-loop automation), defined by a rubric of decision authority, operator role, and the validation evidence each level requires. A second axis records where each study’s performance was demonstrated, from simulation to closed-loop use. Applications include leak and anomaly detection, defect and corrosion assessment, hard spots, pump cavitation, predictive maintenance, failure-risk modeling, scheduling, and digital twins. Cross-classifying the coded sources on both axes shows that validation realism does not keep pace with claimed operational responsibility. Peer-reviewed evaluations are concentrated at Levels 1–3 and rely largely on laboratory, simulated, or retrospective data; the only operational Level 4 example is a practitioner case, and no coded source demonstrates Level 5 closed-loop control. The corpus thus shows a validation gap at the lower levels and an evidence-coverage gap at the upper ones; beyond the performance shown in controlled evaluations, validation, transferability, and governance remain major unresolved barriers to deployment. A ranked agenda led by graduated field validation follows.
- Open Access
Sequence-Preserving Dual-FoV Defense for Traffic Sign and Light Recognition in Autonomous Vehicles
For Autonomous Vehicles (AVs), recognizing traffic lights and signs is critical for safety because perception errors directly affect navigation decisions. Real-world disturbances such as glare, rain, dirt, and graffiti, as well as digital adversarial attacks, can lead to dangerous misclassifications. Current research lacks (i) temporal continuity (stable detection across consecutive frames to prevent flickering misclassifications), (ii) multi-field-of-view (FoV) sensing, and (iii) integrated defenses against both digital and natural degradation. This paper presents two principal contributions: (1) a three-layer defense framework integrating feature squeezing, inference-time temperature scaling (softmax = 3 without distillation training), and entropy-based anomaly detection with sequence-level temporal voting; (2) a 500 sequence dual-FoV benchmark (30k base frames, 150k with perturbations) from aiMotive, Waymo, Udacity, and Texas sources across four operational design domains. The unified defense stack achieves 79.8% mAP on a 100-sequence test set (6k base frames, 30k with perturbations), reducing attack success rate from 37.4% to 18.2% (51% reduction) and high-risk misclassifications by 32%. Cross-FoV validation and temporal voting enhance stability under lighting changes (+3.5% mAP) and occlusions (+2.7% mAP). Defense improvements (+9.5–9.6% mAP) remain consistent across native 3D (aiMotive, Waymo) and projected 2D (Udacity, Texas) annotations. Preliminary recapture experiments (n = 15 scenarios) show 2.5% synthetic–physical ASR gap (p = 0.18), though larger validation is needed. Code, models, and dataset reconstruction tools are publicly available.Open AccessSequence-Preserving Dual-FoV Defense for Traffic Sign and Light Recognition in Autonomous Vehicles
For Autonomous Vehicles (AVs), recognizing traffic lights and signs is critical for safety because perception errors directly affect navigation decisions. Real-world disturbances such as glare, rain, dirt, and graffiti, as well as digital adversarial attacks, can lead to dangerous misclassifications. Current research lacks (i) temporal continuity (stable detection across consecutive frames to prevent flickering misclassifications), (ii) multi-field-of-view (FoV) sensing, and (iii) integrated defenses against both digital and natural degradation. This paper presents two principal contributions: (1) a three-layer defense framework integrating feature squeezing, inference-time temperature scaling (softmax = 3 without distillation training), and entropy-based anomaly detection with sequence-level temporal voting; (2) a 500 sequence dual-FoV benchmark (30k base frames, 150k with perturbations) from aiMotive, Waymo, Udacity, and Texas sources across four operational design domains. The unified defense stack achieves 79.8% mAP on a 100-sequence test set (6k base frames, 30k with perturbations), reducing attack success rate from 37.4% to 18.2% (51% reduction) and high-risk misclassifications by 32%. Cross-FoV validation and temporal voting enhance stability under lighting changes (+3.5% mAP) and occlusions (+2.7% mAP). Defense improvements (+9.5–9.6% mAP) remain consistent across native 3D (aiMotive, Waymo) and projected 2D (Udacity, Texas) annotations. Preliminary recapture experiments (n = 15 scenarios) show 2.5% synthetic–physical ASR gap (p = 0.18), though larger validation is needed. Code, models, and dataset reconstruction tools are publicly available. - Open Access
- 27 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Simulation experiments on the feasibility of replacing human observers in E-VLOS UAV flights with Unoccupied Ground Vehicle units
Extended Visual Line of Sight (E-VLOS) operations are increasingly viewed as a practical intermediate step toward scalable Beyond Visual Line of Sight (BVLOS) unoccupied aerial vehicle (UAV) deployments. However, current E-VLOS practices rely heavily on human observers, introducing limitations in scalability, cost, consistency, and measurable performance assurance. This study investigates the feasibility of replacing human E-VLOS observers with an autonomous unoccupied ground vehicle (UGV) unit equipped with vision-based perception. A heterogeneous UAV-UGV simulation framework is shown in which a visionequipped UGV autonomously tracks and observes a UAV using only monocular visual feedback, without access to UAV state information or telemetry. The system is implemented in a physicsbased simulation environment. It employs a closed-loop visionbased control strategy to maintain continuous visual contact, regulate relative distance, and recover from temporary loss of visibility. Performance is evaluated using metrics directly aligned with E-VLOS observer responsibilities, including visual contact persistence, tracking stability, distance regulation, robustness to target loss, and control effort. Experimental results demonstrate that the UGV maintains persistent situational awareness, exhibits stable and conservative observer behavior, and reliably reacquires the UAV following brief visibility disruptions. The findings suggest that autonomous ground observers can fulfill key functional roles of human E-VLOS participants while offering improved scalability, repeatability, and quantitative performance verification. This work presents a structured simulation pipeline and empirical evidence supporting heterogeneous robotic observers as a viable pathway toward resilient, certifiable BVLOS UAV operations.
- 27 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessSimulation experiments on the feasibility of replacing human observers in E-VLOS UAV flights with Unoccupied Ground Vehicle units
Extended Visual Line of Sight (E-VLOS) operations are increasingly viewed as a practical intermediate step toward scalable Beyond Visual Line of Sight (BVLOS) unoccupied aerial vehicle (UAV) deployments. However, current E-VLOS practices rely heavily on human observers, introducing limitations in scalability, cost, consistency, and measurable performance assurance. This study investigates the feasibility of replacing human E-VLOS observers with an autonomous unoccupied ground vehicle (UGV) unit equipped with vision-based perception. A heterogeneous UAV-UGV simulation framework is shown in which a visionequipped UGV autonomously tracks and observes a UAV using only monocular visual feedback, without access to UAV state information or telemetry. The system is implemented in a physicsbased simulation environment. It employs a closed-loop visionbased control strategy to maintain continuous visual contact, regulate relative distance, and recover from temporary loss of visibility. Performance is evaluated using metrics directly aligned with E-VLOS observer responsibilities, including visual contact persistence, tracking stability, distance regulation, robustness to target loss, and control effort. Experimental results demonstrate that the UGV maintains persistent situational awareness, exhibits stable and conservative observer behavior, and reliably reacquires the UAV following brief visibility disruptions. The findings suggest that autonomous ground observers can fulfill key functional roles of human E-VLOS participants while offering improved scalability, repeatability, and quantitative performance verification. This work presents a structured simulation pipeline and empirical evidence supporting heterogeneous robotic observers as a viable pathway toward resilient, certifiable BVLOS UAV operations.
- Open Access
- 13 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Resource-Aware Adaptive Monitoring for Smart Grid Integrity: A Comparative Analysis of Sparse and Sentinel-Augmented Sensing modes
Continuous integrity monitoring of smart power grids is increasingly challenged by sensing, communication, and operational constraints that limit full observability. While many existing approaches assume dense measurement availability, practical deployments must balance monitoring costs with the timely detection of abnormal grid behavior. This study presents a resource-aware adaptive monitoring framework that explicitly models sensing as a constrained resource and evaluates how partial observability affects detection latency and reliability. Using a MATPOWER-based time-series simulation, two monitoring modes are compared: adaptive threshold-based monitoring and sentinel augmentation. The proposed adaptive strategy dynamically selects subsets of buses for observation and escalates sensing coverage in response to statistically significant voltage deviations. Experimental results demonstrate that pure adaptive monitoring can reduce sensing costs by more than 80% relative to always-on monitoring, albeit with non-negligible detection latency due to sparse observations. Introducing a small set of continuously monitored sentinel buses substantially reduces detection latency, often to zero, at the expense of increased sensing cost and false positives. Parameter sweep analyses further reveal non-monotonic cost-latency behavior, highlighting diminishing returns from uniform increases in sampling density. Rather than identifying a single optimal configuration, this study characterizes the fundamental trade-offs among cost, latency, and sensitivity in resource-constrained smart grid monitoring and provides practical insights for designing flexible integrity-monitoring strategies under limited observability.
- 13 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessResource-Aware Adaptive Monitoring for Smart Grid Integrity: A Comparative Analysis of Sparse and Sentinel-Augmented Sensing modes
Continuous integrity monitoring of smart power grids is increasingly challenged by sensing, communication, and operational constraints that limit full observability. While many existing approaches assume dense measurement availability, practical deployments must balance monitoring costs with the timely detection of abnormal grid behavior. This study presents a resource-aware adaptive monitoring framework that explicitly models sensing as a constrained resource and evaluates how partial observability affects detection latency and reliability. Using a MATPOWER-based time-series simulation, two monitoring modes are compared: adaptive threshold-based monitoring and sentinel augmentation. The proposed adaptive strategy dynamically selects subsets of buses for observation and escalates sensing coverage in response to statistically significant voltage deviations. Experimental results demonstrate that pure adaptive monitoring can reduce sensing costs by more than 80% relative to always-on monitoring, albeit with non-negligible detection latency due to sparse observations. Introducing a small set of continuously monitored sentinel buses substantially reduces detection latency, often to zero, at the expense of increased sensing cost and false positives. Parameter sweep analyses further reveal non-monotonic cost-latency behavior, highlighting diminishing returns from uniform increases in sampling density. Rather than identifying a single optimal configuration, this study characterizes the fundamental trade-offs among cost, latency, and sensitivity in resource-constrained smart grid monitoring and provides practical insights for designing flexible integrity-monitoring strategies under limited observability.
- Open Access
Sequence-Preserving Dual-FoV Defense for Traffic Sign and Light Recognition in Autonomous Vehicles
Traffic light and sign recognition are key for Autonomous Vehicles (AVs) because perception mistakes directly influence navigation and safety. In addition to digital adversarial attacks, models are vulnerable to existing perturbations (glare, rain, dirt, or graffiti), which could lead to dangerous misclassifications. The current work lacks consideration of temporal continuity, multistatic field-of-view (FoV) sensing, and robustness to both digital and natural degradation. This study proposes a dual FoV, sequence-preserving robustness framework for traffic lights and signs in the USA based on a multi-source dataset built on aiMotive, Udacity, Waymo, and self-recorded videos from the region of Texas. Mid and long-term sequences of RGB images are temporally aligned for four operational design domains (ODDs): highway, night, rainy, and urban. Over a series of experiments on a real-life application of anomaly detection, this study outlines a unified three-layer defense stack framework that incorporates feature squeezing, defensive distillation, and entropy-based anomaly detection, as well as sequence-wise temporal voting for further enhancement. The evaluation measures included accuracy, attack success rate (ASR), risk-weighted misclassification severity, and confidence stability. Physical transferability was confirmed using probes for recapture. The results showed that the Unified Defense Stack achieved 79.8mAP and reduced the ASR to 18.2%, which is superior to YOLOv8, YOLOv9, and BEVFormer, while reducing the high-risk misclassification to 32%.
Open AccessSequence-Preserving Dual-FoV Defense for Traffic Sign and Light Recognition in Autonomous Vehicles
Traffic light and sign recognition are key for Autonomous Vehicles (AVs) because perception mistakes directly influence navigation and safety. In addition to digital adversarial attacks, models are vulnerable to existing perturbations (glare, rain, dirt, or graffiti), which could lead to dangerous misclassifications. The current work lacks consideration of temporal continuity, multistatic field-of-view (FoV) sensing, and robustness to both digital and natural degradation. This study proposes a dual FoV, sequence-preserving robustness framework for traffic lights and signs in the USA based on a multi-source dataset built on aiMotive, Udacity, Waymo, and self-recorded videos from the region of Texas. Mid and long-term sequences of RGB images are temporally aligned for four operational design domains (ODDs): highway, night, rainy, and urban. Over a series of experiments on a real-life application of anomaly detection, this study outlines a unified three-layer defense stack framework that incorporates feature squeezing, defensive distillation, and entropy-based anomaly detection, as well as sequence-wise temporal voting for further enhancement. The evaluation measures included accuracy, attack success rate (ASR), risk-weighted misclassification severity, and confidence stability. Physical transferability was confirmed using probes for recapture. The results showed that the Unified Defense Stack achieved 79.8mAP and reduced the ASR to 18.2%, which is superior to YOLOv8, YOLOv9, and BEVFormer, while reducing the high-risk misclassification to 32%.
- Open Access
- 11 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Towards Personal Aerial Vehicles for Urban Air Mobility Transportation: Examining Considerations and Potential Challenges
Personal Aerial Vehicles (PAVs) are poised to reshape urban transportation by offering flexible and on-demand mobility with the potential to reduce traffic congestion, enhance transportation access, and transform urban mobility. This study provides a comprehensive review of emerging trends, technological innovations, regulatory frameworks, and challenges for integrating personal aerial vehicles (PAVs) into modern transportation systems. Through a combination of bibliometric analysis and structured literature review, current research is classified into three core domains: (1) vehicle structure, hardware, and materials; (2) software systems for PAV operations; and (3) policy, regulation, and societal considerations. The classification highlights key trends, innovations, challenges, and technological limitations in each domain. The study also introduces the concept of Aerial Vehicles as a Service (AVaaS) and examines the feasibility of passengers-as-pilots in semi-autonomous operations. Drawing on parallels with Uncrewed Aerial Vehicle (UAV) research, the study identifies three priority areas for advancing PAV integration: (1) developing highfidelity simulation platforms; (2) creating open-access PAV-specific dataset to support data-driven modeling; and (3) implementing robust cybersecurity frameworks to ensure safe, data-driven, and secure operations in urban environments. The insights from this study provide a forward-looking roadmap to guide engineers, policymakers, and infrastructure stakeholders toward the safe, scalable, and socially accepted deployment of PAVs within the Urban Air Mobility (UAM) ecosystem.
- 11 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessTowards Personal Aerial Vehicles for Urban Air Mobility Transportation: Examining Considerations and Potential Challenges
Personal Aerial Vehicles (PAVs) are poised to reshape urban transportation by offering flexible and on-demand mobility with the potential to reduce traffic congestion, enhance transportation access, and transform urban mobility. This study provides a comprehensive review of emerging trends, technological innovations, regulatory frameworks, and challenges for integrating personal aerial vehicles (PAVs) into modern transportation systems. Through a combination of bibliometric analysis and structured literature review, current research is classified into three core domains: (1) vehicle structure, hardware, and materials; (2) software systems for PAV operations; and (3) policy, regulation, and societal considerations. The classification highlights key trends, innovations, challenges, and technological limitations in each domain. The study also introduces the concept of Aerial Vehicles as a Service (AVaaS) and examines the feasibility of passengers-as-pilots in semi-autonomous operations. Drawing on parallels with Uncrewed Aerial Vehicle (UAV) research, the study identifies three priority areas for advancing PAV integration: (1) developing highfidelity simulation platforms; (2) creating open-access PAV-specific dataset to support data-driven modeling; and (3) implementing robust cybersecurity frameworks to ensure safe, data-driven, and secure operations in urban environments. The insights from this study provide a forward-looking roadmap to guide engineers, policymakers, and infrastructure stakeholders toward the safe, scalable, and socially accepted deployment of PAVs within the Urban Air Mobility (UAM) ecosystem.
- Open Access
A workflow for generating synthetic LiDAR datasets in simulation environments
This paper presents a simulation workflow for generating synthetic LiDAR datasets to support autonomous vehicle perception, robotics research, and sensor security analysis. Leveraging the CoppeliaSim simulation environment and its Python API, we integrate time-of-flight LiDAR, image sensors, and two dimensional scanners onto a simulated vehicle platform operating within an urban scenario. The workflow automates data capture, storage, and annotation across multiple formats (PCD, PLY, CSV), producing synchronized multimodal datasets with ground truth pose information. We validate the pipeline by generating large-scale point clouds and corresponding RGB and depth imagery. The study examines potential security vulnerabilities in LiDAR data, such as adversarial point injection and spoofing attacks, and demonstrates how synthetic datasets can facilitate the evaluation of defense strategies. Finally, limitations related to environmental realism, sensor noise modeling, and computational scalability are discussed, and future research directions, such as incorporating weather effects, real-world terrain models, and advanced scanner configurations, are proposed. The workflow provides a versatile, reproducible framework for generating high-fidelity synthetic LiDAR datasets to advance perception research and strengthen sensor security in autonomous systems. Documentation and examples accompany this framework; samples of animated cloud returns and image sensor data can be found at this Link.
Open AccessA workflow for generating synthetic LiDAR datasets in simulation environments
This paper presents a simulation workflow for generating synthetic LiDAR datasets to support autonomous vehicle perception, robotics research, and sensor security analysis. Leveraging the CoppeliaSim simulation environment and its Python API, we integrate time-of-flight LiDAR, image sensors, and two dimensional scanners onto a simulated vehicle platform operating within an urban scenario. The workflow automates data capture, storage, and annotation across multiple formats (PCD, PLY, CSV), producing synchronized multimodal datasets with ground truth pose information. We validate the pipeline by generating large-scale point clouds and corresponding RGB and depth imagery. The study examines potential security vulnerabilities in LiDAR data, such as adversarial point injection and spoofing attacks, and demonstrates how synthetic datasets can facilitate the evaluation of defense strategies. Finally, limitations related to environmental realism, sensor noise modeling, and computational scalability are discussed, and future research directions, such as incorporating weather effects, real-world terrain models, and advanced scanner configurations, are proposed. The workflow provides a versatile, reproducible framework for generating high-fidelity synthetic LiDAR datasets to advance perception research and strengthen sensor security in autonomous systems. Documentation and examples accompany this framework; samples of animated cloud returns and image sensor data can be found at this Link.
- Open Access
- 21 April 2025
- Published by MDPI AG in Preprints.org
A Systematic Review of AI-Backed Tools in Education
Artificial Intelligence (AI) has rapidly transformed education, offering personalization, automation, and learning scalability opportunities. As AI tools gain popularity, there is an immediate need to categorize, evaluate, and understand their varied functionalities and implications within educational environments. This study provides a structured taxonomy of AI-backed tools currently used in education, organizing them into three primary categories: solver tools, multimedia creation tools, and feedback/rephrasing tools. Each category is examined for its functionality and associated challenges, including ethical concerns, data privacy, and potential misuse. The study reviews recent literature and tools to understand methods to reshape the teaching-learning ecosystem. This study intends to inform educators, developers, and policymakers on effective integration strategies and responsible use. It also highlights the importance of AI detection tools, identifies unresolved risks such as AI misinformation and guardrail evasion, and argues for a balanced approach to AI adoption in educational contexts.
- 21 April 2025
- Published by MDPI AG in Preprints.org
Open AccessA Systematic Review of AI-Backed Tools in Education
Artificial Intelligence (AI) has rapidly transformed education, offering personalization, automation, and learning scalability opportunities. As AI tools gain popularity, there is an immediate need to categorize, evaluate, and understand their varied functionalities and implications within educational environments. This study provides a structured taxonomy of AI-backed tools currently used in education, organizing them into three primary categories: solver tools, multimedia creation tools, and feedback/rephrasing tools. Each category is examined for its functionality and associated challenges, including ethical concerns, data privacy, and potential misuse. The study reviews recent literature and tools to understand methods to reshape the teaching-learning ecosystem. This study intends to inform educators, developers, and policymakers on effective integration strategies and responsible use. It also highlights the importance of AI detection tools, identifies unresolved risks such as AI misinformation and guardrail evasion, and argues for a balanced approach to AI adoption in educational contexts.
- Open Access
- 7 March 2025
- Published by MDPI AG in Preprints.org
Drones in Education: A Comparative Analysis of Learning-Focused Platforms
Unoccupied Aerial Vehicles (UAVs) have emerged as valuable educational tools, providing hands-on learning experiences in engineering, programming, and data analysis. This paper presents a comparative analysis of three popular UAV platforms designed for educational purposes: DJI Tello EDU, Robolink CoDrone EDU, and Crazyflie. By evaluating factors such as hardware features, usability, and modularity, this study aims to guide educators and researchers in selecting the most suitable drone platform for their needs. Unique platform features are discussed, and simulation tools, ROS compatibility, and research applications for these drones are also explored. The findings serve as a resource for institutions looking to integrate UAV technology into their programs while balancing cost, functionality, and learning objectives. This evaluative study contributes to the growing field of UAV-based education by offering a structured comparison of drone platforms for enhanced classroom and research applications.
- 7 March 2025
- Published by MDPI AG in Preprints.org
Open AccessDrones in Education: A Comparative Analysis of Learning-Focused Platforms
Unoccupied Aerial Vehicles (UAVs) have emerged as valuable educational tools, providing hands-on learning experiences in engineering, programming, and data analysis. This paper presents a comparative analysis of three popular UAV platforms designed for educational purposes: DJI Tello EDU, Robolink CoDrone EDU, and Crazyflie. By evaluating factors such as hardware features, usability, and modularity, this study aims to guide educators and researchers in selecting the most suitable drone platform for their needs. Unique platform features are discussed, and simulation tools, ROS compatibility, and research applications for these drones are also explored. The findings serve as a resource for institutions looking to integrate UAV technology into their programs while balancing cost, functionality, and learning objectives. This evaluative study contributes to the growing field of UAV-based education by offering a structured comparison of drone platforms for enhanced classroom and research applications.
- Open Access
Integrating Large Language Models for UAV Control in Simulated Environments: A Modular Interaction Approach
The intersection of LLMs (Large Language Models) and UAV (Unoccupied Aerial Vehicles) technology represents a promising field of research with the potential to enhance UAV capabilities significantly. This study explores the application of LLMs in UAV control, focusing on the opportunities for integrating advanced natural language processing into autonomous aerial systems. By enabling UAVs to interpret and respond to natural language commands, LLMs simplify the UAV control and usage, making them accessible to a broader user base and facilitating more intuitive human-machine interactions. The paper discusses several key areas where LLMs can impact UAV technology, including autonomous decision-making, dynamic mission planning, enhanced situational awareness, and improved safety protocols. Through a comprehensive review of current developments and potential future directions, this study aims to highlight how LLMs can transform UAV operations, making them more adaptable, responsive, and efficient in complex environments. A template development framework for integrating LLMs in UAV control is also described. Proof of Concept results that integrate existing LLM models and popular robotic simulation platforms are demonstrated. The findings suggest that while there are substantial technical and ethical challenges to address, integrating LLMs into UAV control holds promising implications for advancing autonomous aerial systems.
Open AccessIntegrating Large Language Models for UAV Control in Simulated Environments: A Modular Interaction Approach
The intersection of LLMs (Large Language Models) and UAV (Unoccupied Aerial Vehicles) technology represents a promising field of research with the potential to enhance UAV capabilities significantly. This study explores the application of LLMs in UAV control, focusing on the opportunities for integrating advanced natural language processing into autonomous aerial systems. By enabling UAVs to interpret and respond to natural language commands, LLMs simplify the UAV control and usage, making them accessible to a broader user base and facilitating more intuitive human-machine interactions. The paper discusses several key areas where LLMs can impact UAV technology, including autonomous decision-making, dynamic mission planning, enhanced situational awareness, and improved safety protocols. Through a comprehensive review of current developments and potential future directions, this study aims to highlight how LLMs can transform UAV operations, making them more adaptable, responsive, and efficient in complex environments. A template development framework for integrating LLMs in UAV control is also described. Proof of Concept results that integrate existing LLM models and popular robotic simulation platforms are demonstrated. The findings suggest that while there are substantial technical and ethical challenges to address, integrating LLMs into UAV control holds promising implications for advancing autonomous aerial systems.