The challenge of search inefficiency arises when multiple UAV swarms conduct dynamic target area searches in unknown environments. The primary sources of this inefficiency are repeated searches in the target region and the dynamic motion of targets. To address this issue, we present the distributed adaptive real-time planning search (DAPSO) technique, which enhances the search efficiency for dynamic targets in uncertain mission situations. To minimize repeated searches, UAVs utilize localized communication for information exchange and dynamically update their situational awareness regarding the mission environment, facilitating collaborative exploration. To mitigate the effects of target mobility, we develop a dynamic mission planning method based on local particle swarm optimization, enabling UAVs to adjust their search trajectories in response to real-time environmental inputs. Finally, we propose a distance-based inter-vehicle collision avoidance strategy to ensure safety during multi-UAV cooperative searches. The experimental findings demonstrate that the proposed DAPSO method significantly outperforms other search strategies regarding the coverage and target detection rates.

To address the limitations of the RRT-series algorithm and its variants, which have demonstrated a lack of dynamic adaptability in various environments, poor path quality, slow convergence rates, and a tendency to become trapped in local optima, this paper aims to propose a dynamic multi-strategy path-planning algorithm based on EG-DRRT* (Evolutionary Game-Theoretic Dynamic RRT*). The integration of evolutionary game theory allows the algorithm to use the concept of dynamically updating replicators to develop the payoff function for a fusion of multi-strategies. This enables a dynamic adjustment of the usage ratios of RRT*, Dijkstra, and Goal Bias algorithms. This approach directs the search process toward converging on the goal point, ultimately achieving an ESS (Evolutionarily Stable Strategy) equilibrium. The experimental results reveal that EG-DRRT* successfully establishes a dynamic balance between global exploration and local optimization across various environments, demonstrating remarkable adaptability and robustness. Additionally, EG-DRRT* shows substantial advantages compared to existing algorithms.

To address the task-grouping problem for air–ground integrated Unmanned Aerial Vehicle (UAV) swarm missions in three-dimensional (3D) environments, this study proposes a data-preprocessing and hybrid initialization clustering method based on 3D spatial features. A dual-modal prototype meta-heuristic optimization model, Dual-Prototype Metaheuristic K-Means (DPM-Kmeans), is constructed accordingly. First, to overcome spatial information loss in high-dimensional task allocation, a 3D spatial task data preprocessing technique and a hybrid initialization strategy based on the golden spiral distribution are designed. This ensures the diversity and environmental adaptability of the initial solutions. Second, a dual-modal prototype optimization framework incorporating row prototypes (local refinement) and column prototypes (global combination) was constructed using meta-heuristics and clustering algorithms. The prototype-driven replacement update mechanism simultaneously performs global and local search, balancing the algorithm’s exploration and exploitation capabilities while expanding the solution space. This effectively addresses premature convergence issues in complex search spaces. Simultaneously, a collaborative multi-constraint, dynamically weighted optimization model was constructed, incorporating task requirements and flight distance constraints to ensure that the grouping scheme approximates the global optimum. Simulation results demonstrate that compared to traditional K-means and mainstream meta-heuristic optimization algorithms, DPM-Kmeans achieves an overall improvement of 2–10% in Sum of Squared Errors (SSE), Silhouette Coefficient (SC), and Davies–Bouldin Index (DB) metrics. It exhibits superior convergence speed and solution quality, proving the method’s excellent scalability and robustness in multi-constraint, large-scale 3D scenarios.

To address the challenges of inefficient convergence in UAV swarms under complex environments due to static position allocation (SPA), as well as the tendency of traditional artificial potential field (APF) obstacle avoidance to get stuck in local optima, this paper proposes a formation control method (DPAF-SA) based on dynamic position allocation (DPA) and APF-SA fusion, grounded in the principle of consensus and the simulated annealing (SA) algorithm. First, the formation position allocation is formulated as an online combinatorial optimization problem. Based on this framework, a dynamic position allocation and dynamic virtual center mechanism is designed to solve the optimal “UAV-position point” mapping in real time, minimizing the total convergence cost of the swarm. Second, to address the local optimum trap and decoupling issues in APF, the global search capability and probabilistic jump mechanism of SA are integrated into APF. This enables optimization of the consistency control input, ensuring tight coupling between efficient obstacle avoidance and formation maintenance. Finally, a high-fidelity HIL simulation platform based on Unity3D 2022.3.2. was established to validate the engineering feasibility and real-time robustness of the proposed algorithm. Simulation results demonstrate that, compared with the representative baseline model, the proposed method achieves improvements of approximately 46.1%, 24.5%, and 39.6% in formation accuracy, convergence performance, and safety margin, respectively, validating its effectiveness.