Purpose A foldable square drone prototype uses the latest structural aerodynamics, energy management and adaptive control to make it practical and reliable for engineering. The design aims to create an unmanned aerial vehicle (UAV) that can be tilted and reconfigured, is lightweight, durable and payload-efficient. Design/methodology/approach The foldable square-frame drone is made of lightweight materials, has a telescopic arm structure and is built in modules to make it more versatile, smaller and last longer. The prototype has a hybrid cargo-handling system, an energy-efficient power system and a foldable frame that makes it easy to carry and improves performance during air transport operations. Findings The foldable square-frame drone worked well in both structural strength and flight stability. The telescopic arm mechanism made it easy to switch between compact and extended configurations, which made the device easier to carry and improved its aerodynamic balance. The drone was able to carry a payload of up to 1.5 kg with very little structural deformation (less than 5 mm). Tests of the flight showed that it could last for up to 25 minutes without a load and 12–15 minutes with one. The dual payload system was able to handle different types of objects while staying stable. Overall, the design struck a good balance between weight, performance and versatility, proving that it was suitable for real-world multi-purpose UAV uses. Research limitations/implications You can switch the drone from compact mode to full-scale mode by pulling out the telescopic arms by hand. In the future, servo-driven actuators will be used to make the transformation process automatic. The multi-section-transformable-UAV model by D'sa and Papanikolopoulos (2019) used hinge actuators to change the shape of the UAV while it was flying. This is similar. Because the drone can fold up, it can go from compact transport mode to operational mode in 40 seconds, which makes it easier to deploy quickly. Practical implications The foldable square drone can be used for many things in everyday life. According to Otto et al. (2018), its structural flexibility makes it easier to deliver lightweight packages, medicines and emergency kits to areas that are far away or have been hit by a disaster. In farming, the drone collects soil samples and moves fertilizer between fields on its own, using Ouyang et al. (2019) UAV-wireless sensor network integration methods. It's great for checking power lines, wind turbines and bridges because it's easy to move and stays put. New studies back up this adaptability. Sheth (2025) and Lee et al. (2025) assert that modular UAV frameworks and morphing-wing technologies enhance mission flexibility and longevity. Social implications The foldable drones are good at balancing portability, endurance and stability. The modular frame can handle mechanical stress when it's full, the distributed energy system keeps the power steady and the hybrid cargo-handling system can handle a wide range of loads. The square foldable drone is a useful, environmentally friendly and flexible choice for both civil and industrial UAV uses, as these results show. Originality/value The proposed folding square-frame drone is unique in its mechanical flexibility, light weight and modularity. It lacks a hard structure like other drones. Instead, its telescopic arms allow it to be packed in a tiny place and fly steadily by modifying rotor spacing. The dual payload system can handle both flexible and rigid loads without tools. Aluminum 6061-T6 is stronger and lighter, making it endure longer and work better. This design works for logistics, farming and emergency response UAVs. Artificial intelligence and renewable energy systems can be added to its scalable platform.
Purpose Wi-Fi channel state information (CSI) sensing is a candidate technology for future Unmanned Aerial Vehicle (UAV)-assisted, device-free human sensing. This paper benchmarks convolutional neural network (CNN) architectures for CSI-based activity and gesture recognition under a controlled, repeated protocol, as a first step towards selecting models for resource-constrained platforms. Design/methodology/approach Two public CSI datasets collected with fixed indoor Wi-Fi infrastructure were used: UT-HAR (activity recognition, 7 classes) and Widar (gesture recognition, 22 classes, body-coordinate velocity profile). Two VGG-inspired CSI-adapted networks (CSI-VGGNet-S and CSI-VGGNet-D), GoogLeNet and DenseNet were compared with four retrained SenseFi baselines (MLP, ResNet-18, LSTM and CNN-GRU). All models were trained with three random seeds under one common budget (at most 20 epochs, early stopping on a held-out validation partition), and the test partitions were used once per run. Sliding-window overlap in UT-HAR was quantified, and recording-level (UT-HAR) and cross-user (Widar) splits were used to probe generalization. No UAV hardware or UAV-collected CSI was used. Findings CSI-VGGNet-D achieved the highest mean test accuracy on UT-HAR (96.80 ± 0.87%) and on Widar (74.43 ± 1.31%). Depth alone did not predict accuracy: GoogLeNet and DenseNet varied strongly across seeds on UT-HAR (SD 16.1 and 12.8 percentage points) despite high training accuracy, and model rankings differed between the two tasks. In the released UT-HAR split, 92% of test windows share at least half of their raw frames with a training window; on a recording-level split without shared frames, the accuracy of CSI-VGGNet-D fell by 6.6 percentage points. Released-split accuracies therefore overstate generalization to unseen recordings, and accuracy, model size and latency must be weighed jointly. Research limitations/implications Results come from fixed indoor infrastructure and a desktop CPU; single-thread latency is only a proxy for embedded hardware, and three seeds give limited statistical power. Validation with UAV-collected CSI and onboard hardware is required before any claim about UAV deployment can be made. Originality/value The paper does not propose a new deep-learning architecture. Its contribution is empirical: a multi-seed, validation-controlled, matched-protocol comparison that re-evaluates SenseFi baselines alongside VGG-style, inception and DenseNet CNNs, quantifies train-test window overlap in the widely used UT-HAR split and its effect on reported accuracy and reports accuracy together with parameters, FLOPs and latency as input to future UAV-oriented model selection.
Purpose Unmanned aerial vehicles are becoming integral across various sectors such as defense, security, industrial, commercial, agriculture and disaster management. Permanent magnet (PM) machines are a crucial component in the propulsion system of UAV due to their high efficiency, power density and compact design. This work proposes the modeling, simulation, performance and harmonic analysis of a PM machine to evaluate its suitability for electric UAV applications. Design/methodology/approach This paper presents the Finite element analysis of 400 V, 40 W and 1000rpm machine. The motor is designed with 18 stator slots and 16 rotor poles using Altair Flux software. The parameters such as mass, phase resistance, inductance, flux density distribution and electromagnetic torque were estimated. Harmonic analysis was performed for magnetomotive force (MMF) winding factor, phase voltage, line-to-line and air-gap flux density to evaluate the machine with respect to vibration and acoustic noise minimization. Findings The PM motor achieved an efficiency of 95.4% with losses of 1.9 W. The phase resistance is 171.65 O, inductance is 0.104 H, mass is 2.449 kg and average electromagnetic torque is 0.37 N-m. The MMF harmonic spectrum has a higher-order harmonics remaining below 0.1 A and the maximum winding factor is 0.95 for the fundamental harmonic. The higher-order harmonics are below 12 V and the flux density of the 5th harmonics is 0.18 T. Originality/value This provides a better way to predict and visualize the performance of the PM motor for electric UAV systems requiring low-vibration, low-noise and energy-efficient performance.
Purpose Reliable operation of power transmission systems depend on timely detection of insulator faults, yet conventional manual inspection is costly, hazardous and inconsistent. This study presents an unmanned aerial vehicles (UAVs)-based vision inspection approach integrating platform development, autonomous flight validation and a controlled multi-generation You Only Look Once (YOLO) benchmark for binary insulator fault detection. Design/methodology/approach A lightweight quadrotor UAV was developed and validated through six autonomous missions along a 132 kV transmission corridor near Prayagraj, India. A dataset of 3,951 annotated insulator images combining 3,173 UAV-collected images and 778 from a publicly available source, comprising 2,847 normal and 1,104 defective instances, was assembled with an 80:10:10 split and flight-level data leakage prevention. Five YOLO variants (YOLOv8s–YOLOv12s) were trained under matched conditions from Common Objects in Context (COCO)-pretrained weights. A FLUKE TiS75+ thermal camera was used separately as complementary qualitative evidence. Findings YOLOv8s achieved the best test-set performance: precision 0.895, recall 0.810, mAP@0.50 of 0.879 and mAP@0.50–0.95 of 0.482 at 142 FPS, outperforming all four newer YOLO generations. Thermal imaging at defective locations revealed hotspot temperatures of 83.5°C–197.6°C against backgrounds below 11°C, corroborating RGB-based fault identifications. Research limitations/implications The dataset was collected under specific environmental conditions near Prayagraj and generalisation to different geographic regions, insulator types and voltage configurations requires further validation. Six autonomous missions represent a limited operational sample. Real-time onboard detection using the Jetson Nano remains future work – all results reported here represent offline evaluation. Thermal imaging was conducted separately and was not fused with the RGB detection pipeline. The binary normal/defective formulation does not distinguish between specific fault types such as contamination, cracking and flashover discolouration, limiting the granularity of fault diagnosis currently achievable by the framework. Practical implications The proposed framework reduces dependence on hazardous manual inspection of overhead transmission lines by enabling systematic, UAV-based image acquisition combined with automated deep learning fault detection. The 142 FPS offline inference speed confirms strong potential for future real-time onboard deployment. The controlled cross-generation YOLO benchmark provides practitioners with an evidence-based model selection guide for similar binary insulator datasets. The complementary thermal imaging results demonstrate that RGB-flagged defective insulators exhibit temperature anomalies consistent with active fault conditions, reinforcing the practical value of multi-modal inspection for power grid maintenance and predictive maintenance programmes. Originality/value This study combines purpose-built UAV development, real-world flight validation, a cross-generation benchmark on a field-collected dataset and complementary thermal corroboration, providing a more complete and reproducible inspection study than single-contribution approaches.
Purpose The purpose is to maintain the confidentiality of UAV-captured images by encrypting them using two novel chaotic map-based dynamic substitution methods. Design/methodology/approach We have proposed two chaotic maps, Coupled-Cosine-Sine (CCS) map and the Amplified-Cosine-Sine (ACS) map. When examined for bifurcation diagrams and Lyapunov exponents, the suggested chaotic maps show a broader continuous chaotic range, complicated behaviour and random distribution. The work also presents an image encryption scheme using the CCS and ACS chaotic maps. The scheme uses permutation and dynamic substitution operations to encrypt the image. The permutation process reorders the pixel positions, while the substitution changes the pixel values. The method employs a two-level permutation approach, operating at both the row and column levels. The dynamic substitution process uses pseudorandom keys and an initial seed, both updated dynamically, CCS, ACS chaotic maps and mathematical operators (XOR and Modulus) to change pixel values. In our scheme, the chaotic map used is not fixed. One of the proposed CCS or ACS maps is used based on the least significant bit of the initial seed; this selection induces the dynamic nature in the scheme. Findings The various result such as UACI, NPCR, BCR etc. are close to their ideal values. Originality/value 1. Proposed the novel CCS map and the ACS chaotic map. 2. Developed a new medical image encryption scheme utilizing these chaotic maps. 3. Introduced a two-level permutation process at the pixel level, comprising row-wise and column-wise transformations. 4. Designed a dynamic substitution mechanism that modifies pixel values based on dynamically updated pseudorandom keys and initial seed, and mathematical operations (XOR and Modulus). 5. Enabled adaptive selection of chaotic maps (CCS or ACS) based on the least significant bit of the initial seed, enhancing the dynamic behaviour of the encryption scheme. 6. A comprehensive security analysis of the proposed system using multiple evaluation metrics is performed to validate its cryptographic strength.
