Advances in Cartography, Mission Planning, Path Search, and Path Following for Drones

1. Introduction

Recently, the rapid advancement and proliferation of drone or unmanned aerial vehicle (UAV) technology have ushered in a new era of possibilities and challenges in various fields. Drones are increasingly employed for applications such as cartography [1,2], surveillance [3,4], delivery services [5,6], environmental monitoring [7,8], agriculture [9,10], data distribution [11], or even emergency and rescue missions [12,13]. They can be remotely operated or programmed to autonomously execute missions, providing cost-effective, flexible, and efficient solutions compared to traditional methods. Their mobility and versatility make them suitable for operations in diverse and often challenging environments.

A wide range of sensors and systems can be integrated into drones to enhance their capabilities. In addition to visual cameras, drones can carry multispectral and hyperspectral imaging devices [14,15], infrared thermal sensors [16,17,18], LiDAR (Light Detection and Ranging) [19,20], and radar systems [21,22]. Furthermore, the integration of sensor fusion techniques, such as Kalman filtering [23,24], dynamic inversion (mathematically invert the known models to determine the control commands necessary to produce a desired response, which is an inverse problem [25]), and machine learning approaches [26,27,28], enables more accurate navigation, positioning, and mission execution. These developments provide a new perspective for research and innovation, while also raising challenges associated with limited payload capacity [29,30], communication constraints [31,32], collision avoidance [33,34], and energy efficiency [35,36,37,38].

Enhancing drone operations also depends on optimizing the computational efficiency of mission planning algorithms. Tasks such as path search, path planning, and path following often involve solving complex optimization problems in real time. Parallelization techniques, including multi-core CPU processing [39,40,41,42] and GPU acceleration [43,44,45], offer promising approaches to significantly increase execution speed and enable faster decision making. By distributing computational loads across multiple processors, drones can achieve more efficient mission execution, especially in time-sensitive applications such as emergency response, collaborative swarms, and dynamic urban environments.

This Special Issue focuses on advanced algorithms and systems for mission planning, including parallelized and distributed approaches for path search, path planning, and path following. The scope extends to collaborative drone systems [46,47], intelligent data acquisition [48], machine learning for pathfinding [49,50], advanced communication [29], and drones in emergency response [51], precision agriculture [9,10], infrastructure inspection [52], and urban planning [53]. The objective is to highlight novel methodologies that improve the efficiency, effectiveness, and safety of drone operations in real-world applications.

This Special Issue brings together contributions from academia, industry, and government agencies, fostering multidisciplinary dialogue on drone technologies. All submitted manuscripts underwent a rigorous peer-review process, ensuring high-quality contributions that reflect the latest innovations in drone navigation and mission planning. The selected works showcase valuable research outcomes and provide insights into the future development of drones, with particular emphasis on their potential societal impact. It is anticipated that subsequent volumes on this topic will continue to broaden the scope, with more in-depth research on autonomous systems, intelligent mission execution, algorithm parallelization, and safety assurance.

2. Statistics of the Special Issue

In total, 22 papers were submitted to this Special Issue. After the peer-review process, 14 papers, including 1 review paper and 13 research articles, representing 63.6% of all submissions, were accepted and published. Six papers, corresponding to 27.3% of the submissions, were rejected during the review process, while 2 papers, accounting for 9.1% of the total, were voluntarily withdrawn by the authors (see Figure 1).

In total, 65 authors contributed to the papers published in this Special Issue, reflecting a broad and dynamic research community dedicated to advancing drone technologies and applications. All contributors participated in a single paper, underscoring the wide reach of the topic and the inclusivity of the Special Issue in attracting a diverse range of perspectives.

Notably, the authors represent an international community, with affiliations from 25 institutes spanning 7 different countries across multiple regions of the world. The geographic distribution of these contributors is presented in

Table 1. Geographic distribution of authors by country.

No.	Country	No. of Authors	Percentage (%)
1	China	46	67.6
2	Croatia	6	8.8
3	South Africa	6	8.8
4	UK	5	7.4
5	New Zealand	3	4.4
6	Australia	1	1.5
7	Canada	1	1.5

and illustrated in Figure 2. This geographic diversity highlights the global significance of drone research, as challenges such as mission planning, navigation, data acquisition, and safety transcend national boundaries and require collective efforts to address. The participation of authors from academia, industry, and governmental institutions further illustrates the multidisciplinary nature of the field, fostering knowledge exchange and innovation. Such international collaboration not only enriches the quality of the contributions but also strengthens the potential for real-world impact, ensuring that the research presented in this Special Issue is relevant across different contexts and environments.

3. Overview of Contributions

This Special Issue brings together papers covering diverse aspects of drone research and development, highlighting advances in mission planning, navigation, multi-UAV coordination, and application-specific solutions.

Zhang S. et al. propose an intelligent loop recommendation method based on a network flow shortest-path model for coordinated multi-UAV operations and introduce a learning-inspired algorithm with Pareto optimization to resolve multi-objective conflicts, providing novel support for intelligent mission management systems.

Lanča L. et al. investigate optimal flight parameters for UAVs equipped with deep-learning-based object detection to maximize search efficiency, analyzing flight speed and ground sampling distance to identify trade-offs between area coverage and detection confidence, offering practical guidelines for reliable mission planning.

Li Y. et al. present a hierarchical task-priority framework to enhance multi-UAV coordination in maritime emergency search and rescue, combining a coverage estimation model, an improved multi-objective gray wolf optimizer for task allocation, and dynamic programming for path planning to reduce task duration and improve coverage of high-value regions.

Tong H. et al. address UAV path planning for complete structural inspection, proposing a mixed-viewpoint generation method and solving a Multi-Layered Angle-Distance Traveling Salesman Problem with a two-step Genetic Algorithm to reduce energy consumption, improve path smoothness, and simplify traversal across varying inspection distances.

Xu T. et al. introduce an energy-optimal trajectory planning method for multi-wing morphing solar UAVs using Hierarchical Reinforcement Learning, enabling adaptive control of thrust, attitude, and wing deflection to maximize energy acquisition, extend endurance, and enhance battery efficiency.

Rahman M. et al. review and classify multi-UAV path planning algorithms from the past eight years, comparing performance across multiple metrics and highlighting open research challenges to guide the development of adaptive and efficient next-generation UAV networks.

Kapari M. et al. apply UAV-acquired spectral data and random forest regression to estimate the maize crop water stress index (CWSI) across vegetative and reproductive growth stages, demonstrating high prediction accuracy and emphasizing UAVs as essential tools for early detection and intervention in smallholder croplands.

Jiang C. et al. develop a 5G UAV inspection path planning method for offshore wind farms, modeling the problem as an obstacle-avoidance traveling salesman problem and proposing a Sea Wind-Aware Improved A*-Guided Genetic Algorithm (SWA-IAGA) to enable efficient and accurate flight paths under complex environmental conditions.

Huang Y. et al. propose an improved sampling-based path planning algorithm, Bi-APF-RRT*, which integrates artificial potential fields, dynamic step sizing, and adaptive sampling strategies to enhance UAV path planning efficiency, obstacle avoidance, and convergence in complex 3D environments, achieving significant reductions in computational time and path length.

Huang Y. et al. also introduce a systematic UAV framework for autonomous exploration of large-scale unknown environments, featuring a low-memory environmental representation and hierarchical planning approach to improve exploration efficiency, reduce memory usage, and enable fast, high-quality path planning, validated through simulations and real-world experiments.

Liu X. et al. propose a distributed consultative temporal consistency guidance law for UAV swarms, integrating dynamic inversion, improved artificial potential fields, and predictive correction to enable synchronized arrival under complex threat conditions, validated through simulation studies.

Wei D. et al. present the OC-MAPPO method, combining optimal control and multi-agent reinforcement learning with GPU parallelization, to enable UAVs to efficiently search for unknown dynamic targets, achieving faster learning and a 26.97% higher success rate compared to Genetic Algorithms.

Esfahlani S. et al. propose an end-to-end UAV logistics framework for emergency medical deliveries, combining hybrid obstacle-aware route planning with time-window-aware scheduling to enable sub-40 min county-wide operations with fewer than ten drones, validated through digital-twin simulations under real-world regulatory constraints.

Sikora V. and Papić T. provide a survey on path planning for multiple moving-object inspection using aerial drones, reviewing state-of-the-art algorithms for motion prediction, obstacle avoidance, and dynamic multi-agent environments, and discussing heuristic, learning-based, and probabilistic approaches as well as future research directions.