Unmanned Aerial Vehicles (UAVs) seem to be the most efficient way of achieving the intended aerial tasks, according to recent improvements. Various researchers from across the world have studied a variety of UAV formations and path planning methodologies. However, when unexpected obstacles arise during a collective flight, path planning might get complicated. The study needs to employ hybrid algorithms of bio-inspired computations to address path planning issues with more stability and speed. In this article, two hybrid models of Ant Colony Optimization were compared with respect to convergence time, i.e., the Max-Min Ant Colony Optimization approach in conjunction with the Differential Evolution and Cauchy mutation operators. Each algorithm was run on a UAV and traveled a predetermined path to evaluate its approach. In terms of the route taken and convergence time, the simulation results suggest that the MMACO-DE technique outperforms the MMACO-CM approach.
This paper addresses the path planning and control of multiple colonies/clusters that have unmanned aerial vehicles (UAV) which make a network in a hazardous environment. To solve the aforementioned issue, we design a new and novel hybrid algorithm. As seen in the mission requirement, to combine the Maximum-Minimum ant colony optimization (ACO) with Vicsek based multi-agent system (MAS) to make an Artificially Intelligent (AI) scheme. In order to control and manage the different colonies, UAVs make a form of a network. The designed method overcomes the deficiencies of existing algorithms related to controlling and synchronizing the information globally. Furthermore, our designed architecture bounds, lemmatizes the pheromone, and finds the best ants which then make the most optimized path. The key contribution of this study is to merge two unique algorithms into a hybrid algorithm that has superior performance than both algorithms operating separately. Another contribution of the designed method is the ability to increase the number of individual agents inside the colony or the number of colonies with a good convergence rate. Lastly, we also compared the simulation results with the non-dominated sorting genetic algorithm II (NSGA-II) in order to prove the designed algorithm has a better convergence rate.
This research offers an improved method for the self-organization of a swarm of UAVs based on a social learning approach. To start, we use three different colonies and three best members i.e., unmanned aerial vehicles (UAVs) randomly placed in the colonies. This study uses max-min ant colony optimization (MMACO) in conjunction with social learning mechanism to plan the optimized path for an individual colony. Hereinafter, the multi-agent system (MAS) chooses the most optimal UAV as the leader of each colony and the remaining UAVs as agents, which helps to organize the randomly positioned UAVs into three different formations. Afterward, the algorithm synchronizes and connects the three colonies into a swarm and controls it using dynamic leader selection. The major contribution of this study is to hybridize two different approaches to produce a more optimized, efficient, and effective strategy. The results verify that the proposed algorithm completes the given objectives. This study also compares the designed method with the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to prove that our method offers better convergence and reaches the target using a shorter route than NSGA-II.
2 December 2024 in Pakistan Journal of Engineering, Technology and Science
In today’s era, facial recognition is common in every aspect of life whether in smartphones, security, attendance management, healthcare, and law enforcement. Modern facial recognition uses a state-of-the-art camera mounted on each device, but we need to integrate it at a certain height for security purposes. To solve this problem, this article proposes real-time facial recognition through drone imaging for urban safety and surveillance. In this article, the Unmanned Aerial Vehicle is equipped with sensors and is allowed to fly in urban locations and capture images to find the suspected one through the image processing technique Convolutional Neural Networks (CNNs). In this research, performances of three facial recognition systems, Local Binary Pattern Histogram (LBPH), FaceNet, and Face_Recognition systems have been presented for Frames Per Second, Accuracy, and real-time tracking. A comparative analysis of the proposed work is also presented to validate the work where a minimum confidence threshold of 80% is achieved.