Drones

UAV pursuit–evasion in multi-obstacle environments constitutes a safety-critical decision-making problem in which the pursuer must intercept a maneuvering evader while satisfying obstacle-avoidance constraints. Existing safety-constrained methods typically optimize expected safety costs or apply immediate interventions, but they do not explicitly determine when recovery should begin once finite-horizon risk emerges. To address this issue, we propose the ReSwitch framework, which separates task-oriented pursuit from safety-oriented recovery. When flight risk emerges, a value-preserving switch-time planner evaluates candidate switching times through opponent-conditioned hybrid rollouts. The feasible switch time with the largest pursuit value after recovery is selected for execution, allowing the controller to prioritize safety while preserving pursuit effectiveness whenever possible. Experiments demonstrate that ReSwitch achieves a mean success rate of 89.02% and a physical safety rate of 91.67%, showing favorable performance compared with the baselines. These results indicate that ReSwitch provides a favorable pursuit–safety trade-off under the tested conditions.

Accurate medium-horizon trajectory prediction, spanning roughly one to three seconds ahead, is essential for the safety and autonomy of uncrewed aerial vehicle (UAV) systems. Trajectory prediction assists UAVs in performing collision avoidance, path planning, and cooperative airspace coordination. Deep sequence models are now widely used for trajectory prediction. The gated recurrent unit (GRU) and long short-term memory (LSTM) models are among the most widely used architectures. However, published comparisons of GRU and LSTM encoder–decoder models rarely use identical data, splits, and hyperparameter budgets. This makes it difficult to attribute reported accuracy differences to the recurrent cell itself. We present a controlled comparison in which two otherwise identical velocity-based encoder–decoder networks jointly predict three future 3D velocity waypoints at +10, +20, and +30 steps ahead. Both networks share an identical 5-fold cross-validation protocol, held-out test split, fixed seed, and 81-point hyperparameter grid, for 405 runs per architecture and 810 runs total. At each architecture’s best configuration, the LSTM model reaches 7.2% lower validation Mean Squared Error (MSE) than GRU. On the held-out test set, LSTM achieves 6.5% lower test MSE under each architecture’s independently selected best configuration (best-practice comparison); a complementary matched-configuration comparison, in which each architecture is retrained under the other’s configuration, shows this advantage is concentrated in robustness to hyperparameter choice rather than in the recurrent cell alone. We also report per-waypoint, per-dimension, and Monte Carlo dropout (MC-dropout) epistemic-uncertainty metrics for both architectures. These metrics are pooled over test windows dominated by synthetic, near-planar flight data (~74%) and should not be read as general conclusions for real, free-form UAV flights. Vertical-velocity error is consistently higher than horizontal-velocity error for both models, reflecting limitations in how well the training data represent vertical movement. The MC-dropout epistemic-uncertainty intervals show 34–36% empirical coverage, substantially lower than the nominal Gaussian-reference target of 68.3% at 1σ. On our 8 × H200 GPU cluster, LSTM’s training time is 51% longer than GRU’s per run, a training-side cost that should not be read as a proxy for embedded inference cost. These results show that LSTM’s main advantage over GRU is its greater robustness to suboptimal hyperparameter settings, rather than substantially better performance when both architectures are well tuned. They also show that the uncertainty estimates from both architectures require post hoc calibration before they can be reliably used as safety bounds.

The growing demand for precise unmanned aerial vehicle (UAV) operations in dynamic environments is often compromised by unmodeled wind disturbances, calling for robust and adaptive control strategies to ensure accurate trajectory tracking. This paper presents a meta-learning augmented model predictive control (ML-MPC) framework for quadrotor trajectory tracking under strong and horizontal wind disturbances with different nominal wind-speed settings. The framework uses a meta-learned basis function to capture shared nonlinear features of aerodynamic disturbances across different wind-speed conditions, while an online adaptation mechanism continuously estimates the corresponding coefficients from flight data. Their combination provides a real-time estimate of the residual aerodynamic force, which is incorporated into the MPC prediction model to compensate for wind disturbances. Extensive flight experiments validate the effectiveness of the ML-MPC framework, showing consistent gains in tracking accuracy across multiple trajectory types and under nominal wind-speed settings of up to , defined by measurements $1\mathrm{m}$ downstream of the fan array. Compared to a baseline GP-MPC controller, the approach achieves average performance improvements of 73.5% in simulation and 50.6% in real-world flight tests, with maximum reductions of 86.5% and 59.7%, respectively.