Unmanned aerial vehicle (UAV) swarms, equipped for distributed sensing and rapid response, can form coalitions to undertake complex missions such as post-disaster relief, communication support, and payload delivery. However, typical coalition formation methods assign each UAV to a single task, limiting cross-task resource sharing. To address this, we investigate overlapping coalition formation (OCF) for UAV swarms, where a single UAV is permitted to participate in multiple coalitions, enabling resource reuse and reducing idleness. We formulate OCF as a multi-objective combinatorial optimization problem that jointly balances task fulfillment ratio, coalition synchronization deviation, and operational cost, while explicitly accounting for inter-coalition resource contention and execution precedence. Specifically, we first construct a hypergraph representation of UAVs and tasks and employ a hypergraph attention network to capture their high-order interactions. Next, we propose a structure-aware hierarchical value decomposition method for policy learning, which progressively aggregates individual- and coalition-level information, models member complementarity and inter-coalition cooperative–competitive relations, and generates a global value estimate that is sensitive to changes in coalition structure. Furthermore, we integrate Monte Carlo Tree Search, utilizing the learned value as a heuristic to efficiently explore the feasible region, and close the loop with candidate-structure demonstration replay and policy distillation, enabling search to refine the learned policy. In multi-scale rescue simulations, the proposed approach improves task utility by up to 11.4% over the best-performing baseline and increases energy efficiency by more than 228% compared to a non-overlapping coalition variant.

In partially observable adversarial environments, heterogeneous unmanned aerial vehicle (UAV) swarms must generate tasks online from noisy observations while respecting platform capabilities, consumable resources, and structural dependencies among tasks. This paper proposes a task graph generation method that converts local observations, target beliefs, and UAV resource states into executable task graphs with explicit resource semantics and inter-task relations. The method first constructs a sufficiently expressive candidate task graph in the belief and resource spaces. An offline search teacher then evaluates future trajectory particles, resource feasibility, and structural interaction values to produce supervision for node selection, marginal task value, and relation prediction. A relation-biased graph attention network learns to generate task graphs online, and a task manager further performs task filtering, dependency repair, conflict completion, and resource checking. Simulation results under complex observation pressure and unseen adversarial strategies show that the proposed method consistently improves structural generation quality and execution feasibility. Compared with Graphormer, it improves the task-graph utility, task-edge F1-score, and executable-graph ratio by 5.83%, 5.41%, and 2.68%, respectively, while reducing the infeasible-task ratio by 35.14%. These results indicate that combining an offline search teacher with resource-constrained graph modeling provides an effective front-end task organization mechanism for heterogeneous UAV swarm planning.

Unmanned aerial vehicle (UAV) swarms exhibit distributed sensing and flexible formation capabilities, which facilitate sustained operations and rapid responses in high-risk, large-area tasks. As a result, they are widely used in disaster relief and related applications. In this paper, we focus on the autonomous decision-making of heterogeneous UAV swarms for accomplishing previously untrained tasks in complex, dynamically evolving post-disaster environments. The swarm's action space typically consists of heterogeneous sub-action spaces corresponding to distinct UAV functions and task modalities, which renders direct policy transfer unreliable or inefficient. Therefore, we propose a multi-agent deep reinforcement learning method for action spaces that are heterogeneous and temporally dynamic. First, a multi-round variational autoencoder is introduced to provide a unified representation of all sub-action spaces within heterogeneous UAV swarms, which mitigates heterogeneity at the representation level. Then, a graph attention network is employed to capture correlations between the represented sub-action spaces, enabling policy transfer to unseen action spaces. However, we find that environmental nonstationarity renders reward signals markedly sparse and delayed, which makes credit assignment difficult. Consequently, we introduce a dynamic hindsight experience replay mechanism that relabels failed trajectories, converting hard-to-use negative samples into informative pseudo-successful samples and thereby significantly improving sample efficiency and training stability. Simulation results in disaster relief scenarios demonstrate that the proposed algorithm generalizes learned policies robustly to unseen action spaces in environments with varying complexities, while exhibiting enhanced resilience under dynamically evolving interference conditions.

Cooperative multi-UAV pursuit–evasion under occlusions and sensor noise is challenged by intermittent observability of the evader, varying observation-window lengths, and non-stationary evader tactics, all of which destabilize prediction and undermine safety-constrained cooperation. To address these challenges, we propose a safe decision-making framework that uses behavior mode and subgoal inference as intermediate representations for interpretable, uncertainty-aware cooperation. Specifically, an observation-driven generative intent–subgoal model infers the evader’s behavior mode and subgoal from short observation windows. Building on this model, a length-agnostic trajectory predictor is trained via multi-window knowledge distillation and consistency regularization to produce future trajectory predictions with calibrated uncertainty for arbitrary observation-window lengths, thereby reducing cross-window inference inconsistency and lowering online computational cost. Based on these predictions, we derive belief and risk features and develop a belief–risk-gated hierarchical multi-agent policy based on soft actor-critic with a safety projection layer, enabling adaptive strategy switching and a controllable trade-off between efficiency and safety. Experiments in obstacle-rich pursuit–evasion environments with randomized layouts and diverse obstacle configurations demonstrate more stable cooperative capture, safer maneuvering, and lower decision variance than representative baselines, indicating strong robustness and real-time feasibility. Specifically, across different observation-window settings, the proposed method improves the normalized expected return by approximately 5–7% over the strongest baseline and reduces pursuer losses by roughly 22–25%. Moreover, its end-to-end decision latency consistently remains within the 50 ms control cycle.