Solving the multi-UAV task allocation problem with complex constraints (MTAPCc) by means of the constrained multi-objective evolutionary algorithms (cMOEAs) is novel research in the field of Operation Research. Its advantages mainly consist of two aspects. One is that it can find feasible solutions that satisfy the constraints within an acceptable time. The other is that the obtained Pareto solution set can offer more options for decision-makers. This paper presents a dualneighborhood search based dual-population coevolutionary algorithm (ccDNCA), which can specifically solve the constrained multi-objective combinatorial optimization problems (cMCOPs) based on permutation encoding, including the MTAPCc. The dual-population coevolutionary framework and the multistrategy collaborative constraint handling method of ccDNCA can effectively improve the efficiency of constraint handling and the ability of finding better solutions. The dual-neighborhood alternating local search (DN-ALS) framework can effectively increase the proportion of feasible solutions during the evolution and enhance the quality of the final solution set. The strategy pool integrated with multiple local search strategies can push the search towards regions with better objective values and constraint values, while enhancing the generalization ability of ccDNCA. In the experimental part, by comprehensively comparing the solution results of ccDNCA with those of other advanced algorithms, it is demonstrated that ccDNCA has significant superiority when dealing with cMCOPs based on permutation encoding, such as the MTAPCc and the Vehicle Routing Problem with Time Window constraints (VRPTW).

The Traffic Alert and Collision Avoidance System (TCAS) is recognized worldwide as the last resort for avoiding midair collisions. The existing TCAS can solve pairwise conflict effectively, but cannot manage multi-aircraft conflict satisfactorily, and more seriously, can even trigger domino conflicts in some situation. In response to the increasingly frequent multi-aircraft conflicts, especially three-aircraft conflicts, it is necessary to improve the ability of TCAS. This paper studies the collision avoidance of multi-aircraft scenarios and innovatively proposes a collaborative optimization of a collision avoidance system (CAS) based on the state prediction of the aircraft. In the process, not only invading aircraft but also potential invading aircraft are considered in the plan for an optimal conflict resolution program. From the perspective of mathematics, the collaborative multi-aircraft conflict detection and resolution algorithm is described in detail in this paper. In the end, this paper conducts a comparative experiment to prove the feasibility of the algorithm in three-aircraft scenarios using InCAS software and Gmas simulation software based on graphical modeling of complex systems. The experimental results show that the CAS proposed in this paper can efficiently prevent the occurrence of domino conflicts and guide each aircraft to avoid conflict areas and return to their origin trajectories. In contrast, the existing TCAS will cause the target aircraft to intensify the conflict with the potential invading aircraft when avoiding intruder aircraft. The research greatly remedies the gaps in the area of multi-aircraft collision avoidance and greatly improves the ability and efficiency of TCAS.

After years of development, low-cost Unmanned Aerial Vehicles (UAVs) play a pyramidally important role in applications of miscellaneous fields by virtues of their excellent mobility and safety. With the increasing difficulty of practical applications, UAV technology research is evolving towards the direction of multi-UAV collaboration. As the primary problem of multi-UAV cooperative mission, path planning attracts more and more researchers' attention. It is followed by diversified multi-UAV path planning algorithms. This paper therefore provides a review of this issue and proposes a novel systematic classification of path planning algorithms. According to this classification, this paper mainly summarizes the typical path planning methods and their corresponding application or development on multiple UAVs. The performance of each algorithm is also discussed. Finally, this paper prospects the future development trend of multi-UAV path planning.

This research presents a novel deep learning framework designed to enable autonomous navigation of unmanned aerial vehicle (UAV) swarms through confined spaces, specifically focusing on the challenge of hole crossing. Leveraging the Gazebo simulation environment, we develop and test a vision-based swarm control strategy that employs onboard cameras and a TensorFlow-based end-to-end controller for decentralized, onboard decision-making. Our approach eliminates the reliance on inter-UAV Wireless communication, instead utilizing deep learning algorithms to process visual inputs and guide UAVs through narrow openings autonomously. Experimental results demonstrate the proposed system's robustness in formation control and obstacle avoidance, achieving successful navigation without the need for external positional data sharing. The study compares the performance of our deep learning-driven model against traditional control systems, revealing comparable efficiency and highlighting the potential for practical applications. This research advances the field of autonomous UAV operations, offering a scalable and effective solution for UAV swarm navigation in complex environments, thereby opening new avenues for the application of UAV swarms in areas such as search and rescue, surveillance, and environmental monitoring.

Multi-UAV autonomous navigation in complex environments presents significant challenges, particularly in communication-denied scenarios where traditional coordination methods fail. Vision-based approaches offer a promising alternative by enabling operation with only onboard sensors, but face fundamental limitations: modular approaches separating perception, mapping, and planning introduce compounding errors across stages, while end-to-end learning methods struggle with sample inefficiency and poor generalization to different environments. We present Imitation and Exploration Learning (IEL), a framework that enables robust communication-free multi-UAV coordination in cluttered environments. Unlike previous methods that treat imitation learning and reinforcement learning as separate paradigms, IEL integrates them through a unified objective function that dynamically balances expert guidance with reinforcement-based exploration. Our implementation combines DAgger's iterative expert supervision with Soft Actor-Critic's maximum-entropy reinforcement learning within a Centralized Training with Decentralized Execution architecture. Extensive experimental validation demonstrates that IEL achieves 95.5% of the performance of privileged-information models while operating solely with onboard visual sensors. Most significantly, IEL maintains 85.4% performance in challenging environments where conventional imitation learning approaches degrade substantially, confirming its effectiveness in bridging the efficiency-generalization gap. Our approach consistently outperforms both pure imitation learning and reinforcement learning baselines across multiple metrics, including collision avoidance, formation maintenance, and adaptive navigation. The principles established in this work extend beyond UAVs to broader multi-agent systems requiring coordination without communication infrastructure. By successfully integrating complementary learning paradigms, IEL contributes to the understanding of robust learning approaches in autonomous systems and establishes a promising direction for resilient multi-agent technologies.

Unmanned Aerial Vehicle (UAV) carries relief materials to perform cooperative delivery mission, which has a high guarantee in both flexibility and security, and can avoid the personal safety threat of pilots in the original delivery mode in the post-disaster scenario. In this paper, we study the problem of collaborative path planning before start of the multi-UAV cooperative delivery mission. Based on the knowledge of graph theory and the generalized multiple traveling salesman problem, the model of multi-UAV mission path planning problem is established. Real and reasonable constraints are added and formulated, including UAV cooperation constraints and task order constraints. According to the characteristics of the problem model, this paper innovatively proposes a Hybrid Grouping Particle Swarm Optimization (HGPSO) algorithm. The particle population is divided into subgroups of the same size to represent the feasible solution. Then, based on the crossover and mutation idea of genetic algorithm, this paper improves the velocity update rule of particle swarm, and proposes four crossover and mutation methods. Finally, simulation experiments verify the effectiveness of the algorithm, and the convergence data show that the algorithm can meet the needs of rapid decision-making and response in post-disaster rescue.