This paper investigates the finite-time position tracking control problem for quad-rotor unmanned aerial vehicles subject to parametric uncertainties and state constraints. Unlike existing works that predominantly focus on ideal quad-rotor systems without physical limitations, unknown parametric uncertainties arising from imprecise mass identification or time-varying aerodynamic disturbances are explicitly considered in this paper. In addition, to ensure flight safety and regulatory compliance, both the position and velocity states are constrained within prescribed compact sets throughout the mission. To address these challenges, an adaptive finite-time control scheme is proposed integrating a logarithmic barrier Lyapunov function within the backstepping-based framework, which simultaneously enforces the prescribed state constraints and compensates for the unknown parametric uncertainties. Rigorous Lyapunov-based stability analysis guarantees that the position tracking errors converge to zero within a finite settling time and all state constraints are strictly satisfied. Simulation results validate the effectiveness of the proposed method. This work advances the state-of-the-art by providing an innovative framework for constrained finite-time control under parametric uncertainties, with potential applications in safety-critical quad-rotor unmanned aerial vehicles operations.
The work intends to study the optimized leader-following consensus control with fault-tolerant ability for a class of second-order nonlinear multi-agent systems (MASs) subjected to actuator failures. Since MAS is an important topic in the field of low-altitude of technology and engineering, this work can contribute to the development of this field. To realize optimized control, reinforcement learning (RL) strategy is employed for avoiding the derivation of analytical solution of Hamilton–Jacobi–Bellman (HJB) equation. Nevertheless, second-order MASs need to simultaneously regulate both position and velocity states to reach consensus, which inevitably increases the complexity of optimization algorithms. In this context, to endow the optimal control scheme with fault-tolerant capability against actuator faults, an adaptive estimation algorithm for unknown actuator fault parameters is integrated with the RL-based optimal control strategy. For making the combination smoothly, a simplified RL algorithm is obtained by taking the negative gradient of a simple positive function, which is equivalent to HJB equation. Finally, both theoretical analysis and numerical simulations verify the feasibility of the proposed control.
To address the degradation of unmanned aerial vehicle (UAV) positioning accuracy caused by variations in sensor performance under changing environments, a scene-matching-based multi-sensor fusion algorithm, termed mixed credibility unscented Kalman filter (MCUKF), is proposed. The method jointly considers environmental and motion scenes for adaptive sensor selection, evaluates data credibility using generalized and narrow Jaccard coefficients, and employs an improved AdaBoost classifier for intelligent scene recognition. By combining filtering credibility and data credibility, the proposed MCUKF enables adaptive fusion of multi-sensor data. Experimental results from simulations and real flight tests demonstrate its effectiveness. In the adaptive sensor fusion experiment, the proposed method achieves an RMSE of 0.1333[Formula: see text]m, compared with 0.6403[Formula: see text]m for the Global Positioning System (GPS), 0.7697[Formula: see text]m for the camera, and 0.4873[Formula: see text]m for direct GPS-camera fusion. In comparison with representative fusion algorithms, MCUKF attains an RMSE of 0.1806[Formula: see text]m, outperforming MMSE (0.5140[Formula: see text]m), EUKF (1.3326[Formula: see text]m), and STETFF (0.9568[Formula: see text]m). These results demonstrate the superiority of the proposed method in adaptive sensor selection and robust fusion during scene transformation.
This paper develops fast algorithms for Nash equilibrium (NE) seeking in multi-group distributed resource allocation games (DRAGs) subject to feasibility-preservation and time-critical requirements for real-time implementation. The considered framework captures hybrid cooperative–competitive interactions among agents, where cooperation arises within groups while conflict exists across groups. Two classes of multi-group DRA games, namely the intra-independent distributed resource allocation game (IIDRAG) and the intra-dependent distributed resource allocation game (IDDRAG), are investigated. Two distributed NE seeking algorithms for both cases are proposed. By exploiting a Laplacian-matrix-based affine transformation, the developed algorithms maintain the hard supply–demand balance of each group, which is desirable for real-time implementation that requires feasibility preservation. Moreover, the sampled-data communication mechanism enables specified-time convergence to the NE while reducing communication burden. It is proved that both algorithms achieve convergence to the NE at a prescribed settling time. Numerical examples further illustrate the effectiveness of the specified-time algorithms and hard supply–demand balance preservation.
This paper proposes a non-parametric automatic tuning method for satellite attitude control that accounts for modeling uncertainties in the inertia and actuator dynamics. The method utilizes a two-relay feedback test, which induces self-sustained periodic oscillations in the system’s output. The existence of the periodic oscillations is analyzed using the describing function method. The profile of these oscillations is used to optimally tune the control parameters through the proposed optimal tuning rules (tuning formulas), with a guarantee on a specific phase margin. The tuning rules are found through numerical optimization for a class of satellites ranging from nanosatellites to large scale satellites, where the attitude dynamics vary depending on the satellite sample. The tuning rules map the oscillations’ amplitude and frequency to the optimal controller gains. The tuning of the controllers happens online in real-time using these tuning rules. The effectiveness of the proposed tuning procedure is evaluated through nonlinear three-axis satellite simulations and validated experimentally using a magnetic-levitation testbed simulating a frictionless space environment.
Intention recognition of non-cooperative space objects is a crucial problem for space situational awareness. This paper proposes a method for intention recognition of close-range space non-cooperative objects before intention realization based on the gated recurrent unit (GRU)-self-attention network (GRU-SA). Typical intention set is defined and the comprehensive learning particle swarm optimization (CLPSO) algorithm is adopted to construct the dataset. The proposed network is trained and tested on the dataset. The results demonstrate the effectiveness of the proposed method in recognizing the intentions of close-range space non-cooperative objects and verify the robustness of the model.
Autonomous operation of Unmanned Aerial Vehicles (UAVs) in complex low-altitude environments requires navigation strategies that ensure safety, adaptability, and reliability under dense obstacle conditions. This paper proposes an end-to-end intelligent navigation framework that integrates a real-time object detection network You Only Lock Once (YOLO) with an adaptive control policy based on Proximal Policy Optimization (PPO) to achieve robust, collision-free obstacle avoidance. The approach is validated in a simulated low-altitude [Formula: see text] 10 m 3 three-dimensional workspace containing 25 static obstacles arranged in a deterministic-stochastic distribution representing highly cluttered operational scenarios. Perception inputs derived from YOLO detections are transformed into a structured observation vector and used to guide continuous control actions learned via a shaped reward function that accounts for goal convergence, altitude stability, motion efficiency, and safe obstacle standoff distances. To better reflect realistic sensing conditions, perceptual uncertainty is modeled by injecting Gaussian noise and false-positive detections into the perception pipeline during training. Experimental results over 500 000 training timesteps demonstrate stable policy convergence, achieving a 90% success rate with zero collisions across multiple test episodes and maintaining consistent performance under varying noise levels. The resulting trajectories show improved safety-aware maneuvering and reduced path length, highlighting the framework’s potential for low-altitude engineering applications requiring dependable autonomous navigation under imperfect perception.
During cooperative hunting, the hawk flock rapidly selects a lead hawk to carry out attacks while other members assist in encircling the prey. Inspired by this, this paper proposes a method for distributed attacker selection in an unmanned aerial vehicle (UAV) swarm. First, incorporating the traits observed in the hawk flock, the fully locally computable situation indicator, individual and group interception indicators, and a feedback-based willingness indicator are designed. Then, the individual utility function and the feasible strategy set are constrained, thereby establishing a distributed selection model that satisfies the conditions of a potential game. Furthermore, a Distributed Strictly-Better Updating (DSU) algorithm is proposed by a strictly-better set and captures the marginal gain, which guarantees convergence within finite steps.
Tilting rotor Unmanned Aerial Vehicles (UAVs) offer great versatility for its capability in vertical take off as a fixed wing aircraft. However, it is considerably more complex in terms of control, especially regarding the changing system dynamics in transition flight. Whereas conventional control methods require accurate mathematical models, reinforcement learning succeeds by leveraging data taken directly from its environments. In this research, we present a reinforcement learning-based control for a tilt rotor UAV in transition flight using the Proximal Policy Optimization (PPO). We propose a special reward function to track altitude and speed during the transition phase to ensure minimal altitude loss and accurate speed hold control. Furthermore, we present the agent with an additional training phase under simulated wind gust and signal delays to improve its robustness. The algorithm uses the built in Reinforcement Learning Toolbox in MATLAB with the tilting rotor UAV environment modeled in Simulink. The agent is trained with randomized initial conditions and reference signals, simulated wind gusts, and signal delays. The results show that the agent is able to achieve accurate speed and altitude tracking. Furthermore, training with the simulated wind gusts does improve the controller performance with regard to the tracking accuracy. Meanwhile, simulated signal delay helps with the smoothness of the produced action. Overall, it is shown that reinforcement learning allows for great versatility with apparent robustness, which offers better performance and flexibility than the traditional tilt rotor UAV control.
The search and containment problem requires a multi-agent system to alternate between dispersion for information acquisition and gathering for target capture under spatiotemporal constraints. Existing studies typically treat search and containment as distinct tasks, whereas this paper shows that they can be addressed within a unified endpoint decision formulation under stage-dependent parameter settings. Specifically, this paper presents a unified framework that predicts the target's position via a hierarchical information map, and generates endpoints in discrete space using a convolution-based greedy algorithm. Then, Gaussian belief propagation is used to refine endpoints and trajectories in continuous space while enforcing safety constraints and stage-dependent termination conditions. Simulations across diverse maps demonstrate order-of-magnitude runtime reductions relative to an integer programming solver, while achieving competitive terminal time and a lower collision rate compared to a baseline that designs separate algorithms for search and containment.
This paper addresses cooperative control and communication optimization for vehicle platoons with random channel loss between the leader and following vehicles. To reduce the communication burden of the system, a distributed control strategy based on an adaptive event-triggered mechanism is proposed.Under a Leader-Predecessor-Following (LPF) communication topology, the random channel loss is modeled as a Markov process, which captures the practical scenario where only a subset of leader-to-follower communication links are available at any given time. A hyperbolic secant function is employed to dynamically adjust the event-triggered threshold. This strategy only triggers information transmission at necessary times, significantly reducing the frequency of data exchange. Subsequently, the vehicle platoon system is formulated as a Markovian switching system with platoon errors serving as state variables. Based on Lyapunov stability theory, sufficient conditions for mean-square asymptotic stability are derived, and a design method for the distributed controller is presented. Finally, the effectiveness of the proposed method under channel-constrained scenarios is validated through simulations.
This paper investigates the performance of a spacecraft equipped with a Sun-facing diffractive sail within a heliocentric mission scenario corresponding to an augmented Hohmann transfer. A diffractive sail is a particular type of solar sail that converts solar radiation pressure into thrust through a large membrane deployed in orbit. However, unlike reflective sails, diffractive sails exploit the properties of a highly engineered thin metamaterial film to generate a thrust vector with specific characteristics that cannot be achieved by a conventional reflective solar sail. Within an optimization framework based on classical variational calculus, this paper analyzes the potential of a Sun-facing diffractive sail with an electro-optically controlled film to minimize the total velocity change required by the augmented Hohmann transfer. Numerical simulations indicate that a sail with a reference propulsive acceleration of about 0.5mm/s(2) enables a reduction of the velocity change of a classical Hohmann transfer by roughly 30% in a simplified Earth-Mars or Earth-Venus scenario.
This paper investigates the prescribed-time trajectory tracking control problem for manipulator systems subject to actuator saturation. To tackle this issue, a bounded time-varying gain function is designed. With its assistance, an adaptive sliding mode control (SMC) strategy is proposed to ensure that the tracking error converges to zero within the prescribed time. Furthermore, a saturation compensation mechanism is developed to handle the actuator saturation. To ensure the feasibility of the present method, sufficient conditions are derived through theoretical analysis to guarantee the prescribed time stability (PTS) of the manipulator system. Finally, numerical simulations are conducted to verify the effectiveness and and superiority of the present control method.
To address resilient distributed consensus of multi-quadrotor unmanned aerial vehicles (QUAVs) in complex low-altitude communication environments, this paper proposes a resilient distributed consensus control strategy for multi-QUAVs subject to limited communication bandwidth and denial-of-service (DoS) attacks. Unlike existing approaches that rely on high communication frequencies or lack explicit mechanisms to handle network attacks, the proposed two-layer control architecture enables secure consensus of multi-QUAVs under constrained communication resources. An adaptive real-time compensation mechanism based on a radial basis function neural network (RBFNN) is incorporated to substantially enhance control accuracy and robustness while suppressing the effects of uncertain nonlinear dynamics. Lyapunov-based stability analysis is employed to rigorously establish asymptotic convergence of the closed-loop system under composite constraints, while systematically addressing the interplay among strong nonlinear dynamics, quantizer constraints, and security issues induced by DoS attacks. Numerical simulations validate the effectiveness of the proposed control strategy and demonstrate superior performance in terms of control accuracy, convergence speed, and robustness against the dual constraints of limited communication bandwidth and DoS attacks.
In the terminal attack phase, aircraft trajectory planning based only on geometric or dynamic optimal solutions often fails to reflect the tactical intent of the mission which limits the overall effectiveness of penetration operations. To address this problem, this paper embeds tactical intent into the path planning process through tactical template decisions. Structured tactical templates are introduced to guide and constrain maneuver behaviors at a semantic level. This enables the generation of multi-aircraft autonomous penetration strategies driven by tactical intent. The penetration success rate is improved by reducing the average flight altitude and exposure time. Based on this idea, a distributed reinforcement learning-based algorithm is proposed for autonomous tactical template decision-making of multiple aircraft. At the lower level, terminal penetration trajectories and corresponding maneuver parameters are generated under tactical template constraints. At the upper level, a multi-agent reinforcement learning method is used to select the optimal combination of tactical templates based on real-time battlefield information. Extensive simulation experiments in various random battlefield scenarios, as well as comparisons with baseline methods, demonstrate that the proposed approach is effective and robust in producing high-quality tactical decisions in complex and dynamic combat environments.
A path planning and control framework is developed for air-ground heterogeneous robots in urban underground utility tunnels. A passable height-interval grid map updated by Extended Kalman filtering provides a unified environment representation for heterogeneous robots. By combining Space-Time Theta* and model predictive control, safe cooperative navigation is achieved.
This paper develops a position-domain guidance optimization framework that adopts along-track position as the independent variable, enabling spatially indexed enforcement of geometric constraints and decoupling trajectory shaping from time parameterization. Building on this framework, a position-domain generalized spectral model predictive convex programming (PGS-MPCP) method is proposed to remove reliance on a prespecified terminal time or fixed time window, thereby mitigating time-window-induced infeasibility in multi-constraint missions involving detours, altitude corridors, and no-fly-zone avoidance. The method is validated on a representative sea-skimming precision-strike planning problem with waypoint and terminal impact-angle constraints. Numerical results demonstrate that PGS-MPCP provides substantial improvements over representative MPSP variants in terms of terminal/waypoint accuracy and feasibility preservation. Relative to convex-optimization and pseudospectral methods, it achieves a favorable balance among computational efficiency, constraint satisfaction, and terminal performance, which makes it promising for online planning and real-time guidance. Additional Monte Carlo studies under both initial-condition perturbations and waypoint-geometry uncertainty further verify stable convergence and robust feasibility preservation.
Rotorcraft aerial robots offer unique advantages in aerial mobility, making them ideal for a wide range of missions. However, the high energy consumption inherent to flight significantly limits their practical deployment. In recent years, air-ground amphibious Unmanned Aerial Vehicles (UAVs) have emerged as a promising solution by integrating ground locomotion capabilities, enabling substantial energy savings and extended endurance while preserving aerial agility and maneuverability. Nevertheless, these multimodal systems pose new control challenges, such as mode transitions between air and ground, dynamic uncertainties during frequent takeoffs and landings, and the complexities of ground contact dynamics. To address these issues in a unified and scalable manner, reinforcement learning offers a compelling alternative to traditional hand-crafted control schemes. This paper focuses on an amphibious UAV with a ducted-fan configuration and develops a high-fidelity simulation environment with contact dynamics. A deep reinforcement learning framework is proposed to achieve robust trajectory tracking across diverse motion states, including aerial flight, ground cruising, takeoff, and landing. Compared with conventional feedback controllers, the trained policy demonstrates superior tracking accuracy. Furthermore, a novel potential field-based reward function — combining attractive terms at both long and short ranges — is introduced to enhance training efficiency and performance. Experimental results show consistent improvements over baseline reward designs across all motion modes, highlighting the framework’s effectiveness and generality.
Accurate three-dimensional (3D) reconstruction of transformers is essential for power system monitoring and digital operation. To address the limitations of existing methods under limited viewpoints, which often lead to incomplete or blurred reconstructions, this paper proposes a limited-view 3D reconstruction framework based on Segment Anything Model (SAM) cross-view fusion and Gaussian diffusion refinement. The method generates a coarse model through cross-view feature fusion and refines geometry and textures using Gaussian diffusion combined with multi-view rendering. Experimental results show that the proposed approach achieves the best or near-best performance across different numbers of views, producing high-quality reconstructions with consistent structures and clear textures even under limited viewpoints. On the public Mip-NeRF 360 Dataset, our method achieves a minimum Learned Perceptual Image Patch Similarity (LPIPS) of 8.97, with Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) improving to 22.36[Formula: see text]dB and 0.9124, respectively, significantly outperforming all other baseline methods.
