Hierarchical coding of space in memory systems and fine-to-coarse wayfinding strategies are used in human navigation. Inspired by this observation, a spatial representation model of multilevel topological route map (MTR-map) and an efficient non-complete route (NC-route) planning algorithm are proposed. First, the model of the MTR-map and its construction algorithm are proposed. The MTR-map model has a hierarchical organization structure. Such an organization structure can be represented by a nested graph in which the nodes of the lower layer are grouped together and form new nodes at the next hierarchical level. For example, places are grouped together to form superordinate regions. Then, based on the MTR-map, a NC-route planning algorithm, called non-complete A* (NC-A*), is proposed, which uses fine spatial information for the close locations and coarse spatial information for distant locations simultaneously. This is a bionic region-based robot route planning strategy, which effectively reduces the size of the search space and improves the search efficiency in the case of sub-optimal route length, and has the advantages of small memory usage, small computational consumption, short planning time and so on. Through experiments, the validity and reliability of the method proposed in this paper are verified.

This article addresses the magnetic source localization problem by developing a receding horizon reinforcement learning approach. First, a Gaussian process regression approach is employed to model the magnetic environment and predict the magnetic strength distribution. Second, the predicted strength is used as the reward function for reinforcement learning. An important characteristic is that the reward function is time-varying as new magnetic data arrive. Third, owing to these time-varying rewards, a receding horizon control strategy is introduced into reinforcement learning to form the receding horizon reinforcement learning approach. The proposed approach not only retains the ability of the receding horizon control approach to deal with time-varying rewards, but also exploits the learning capability of reinforcement learning to adapt to the environment in real time. Finally, simulations and experiments illustrate the effectiveness of this proposed approach that guides a quadrotor to search and locate the magnetic sources.

above article [1], Fig. 21(b) (Flight Controller) contains a description error. In the original figure, the flight controller was described as related to MAVROS. However, the UAV platform used in this study is FanciSwarm, and its flight controller is Mcontroller, which is independently developed by Fancinnov. The flight controller does not directly communicate with ROS through MAVROS; instead, data exchange with ROS is achieved through a self-developed flight controller interface provided by Fancinnov. To avoid confusion and improve reproducibility, the flight controller in Fig. 21(b) has been revised and labeled as Mcontroller-V7, and the communication mechanism with ROS has been clarified accordingly. This correction is intended only to improve the accuracy of the UAV hardware architecture description and does not affect the experimental results, analysis, or conclusions reported in the paper.Fig. 21.Real-world environment. The maximum velocity of each UAV is set as $1.5m/s$ . A remote control car equipped with UWB is used as the dynamic target.

In realistic radar tracking scenarios, the target measurement uncertainty (TMU)—encompassing both detection probability and measurement error covariance—is strongly influenced by the target-to-radar (T2R) geometry. However, the existing posterior Cramér–Rao lower bounds (PCRLBs) largely overlook the fundamental impact of T2R geometry on the TMU and, consequently, on the mean square error (MSE) of state estimation, often leading to overly conservative bounds. To bridge this gap, this paper first develops a generalized model of target measurement error covariance for bistatic radar systems with moving transmitters and receivers, in which the impact of T2R geometry on the error covariance is explicitly characterized. Based upon this TMU formulation, we subsequently derive a geometry-dependent PCRLB (GD-PCRLB) that fully incorporates both measurement origin uncertainty and geometry-dependent TMU. In this derivation, both detection probability and measurement error covariance are treated as state-dependent parameters when differentiating the log-likelihood function with respect to the target state. Unlike existing PCRLBs that partially or completely ignore the geometry-dependent nature of TMU, the proposed GD-PCRLB captures substantial performance benefits by extracting additional Fisher information arising from geometry-dependent TMU. The resulted GD-PCRLB provides a significantly less conservative MSE lower bound compared to existing bounds that only partially or fully neglect the influence of T2R geometry on TMU. Numerical results demonstrate that the improvement offered by the GD-PCRLB becomes more pronounced as the level of TMU increases.

The conditional posterior Cramer-Rao lower bound (CPCRLB), incorporating historical measurements from particular realizations of target states, is regarded as a more accurate and faithful online lower bound compared to the unconditional posterior Cramer-Rao lower bound (PCRLB). In general radar, especially bistatic radar tracking, the target measurement uncertainty (TMU) in terms of both target detection probability and measurement error covariance is significantly determined by the target-to-radar (T2R) geometry. However, the existing CPCRLBs assume perfect target detection and lead to overoptimistic conditional mean-square-error (MSE) lower bounds for radar tracking with miss-detection. Moreover, they completely ignore the impact of geometry-dependent TMU on the conditional bound and inevitably discard valuable target Fisher information, which makes the bounds overconservative. This article rigorously derives a generalized CPCRLB (GCPCRLB) to fully account for the impact of both the miss-detection and geometry-dependent TMU on the conditional bound. Furthermore, we prove that the proposed GCPCRLB coincides with the existing CPCRLB under the assumption of both perfect detection and geometry-independent TMU. Based on the general recursion of the proposed GCPCRLB, an implementable GCPCRLB is further derived for bistatic radar tracking with explicit geometry-dependent TMU. The proposed implementable GCPCRLB is then applied to radar trajectory control to optimize the T2R geometry for improved tracking. Numerical results demonstrate that the proposed GCPCRLB provides a much more accurate conditional MSE lower bound for radar tracking with miss-detection and geometry-dependent TMU. Compared to state-of-the-art radar trajectory control methods, our proposed control method acquires the target measurement with the least uncertainty and achieves the most accurate tracking result.

Highlights What are the main findings? A map-change-driven replanning (MCR) strategy is developed to adaptively trigger trajectory replanning based on ESDF structural variations and goal drift in unknown indoor environments. The proposed closed-loop Autonomous Unmanned Aerial Vehicle (UAV) navigation system achieves higher safety and lower replanning frequency than conventional time-based replanning strategies in cluttered scenarios. What are the implications of the main findings? Linking replanning decisions to real-time map evolution enables more stable and resource-efficient autonomous flight under partial observability. The presented framework demonstrates a practical system-level solution for indoor UAV exploration and inspection tasks under realistic operational constraints.Highlights What are the main findings? A map-change-driven replanning (MCR) strategy is developed to adaptively trigger trajectory replanning based on ESDF structural variations and goal drift in unknown indoor environments. The proposed closed-loop Autonomous Unmanned Aerial Vehicle (UAV) navigation system achieves higher safety and lower replanning frequency than conventional time-based replanning strategies in cluttered scenarios. What are the implications of the main findings? Linking replanning decisions to real-time map evolution enables more stable and resource-efficient autonomous flight under partial observability. The presented framework demonstrates a practical system-level solution for indoor UAV exploration and inspection tasks under realistic operational constraints.Abstract Autonomous Unmanned Aerial Vehicle (UAV) navigation in unknown indoor environments is challenged by incremental map revelation and non-uniform geometric changes, which frequently invalidate preplanned trajectories. Existing time-triggered replanning strategies are poorly aligned with such irregular environmental evolution, often resulting in either redundant computation or delayed responses to critical structural variations. To overcome these limitations, this paper proposes a map-change-driven closed-loop replanning mechanism (MCR) embedded within a distance-field-based hierarchical exploration-planning-control framework. The proposed approach explicitly monitors local Euclidean Signed Distance Field (ESDF) structural changes and exploration goal updates, triggering replanning only when significant geometric or task-level variations are detected. This event-driven design enables timely trajectory adaptation while effectively suppressing unnecessary replanning. Extensive experiments conducted in a high-fidelity indoor warehouse simulation environment demonstrate that the proposed method consistently outperforms single-shot planning and fixed-interval replanning baselines in terms of task success rate, trajectory smoothness, safety margin, and replanning efficiency. These results validate the effectiveness of using map structural evolution as the core driver for replanning in unknown indoor UAV navigation.

This paper deals with the problem of dynamic trajectory planning for a group of unmanned aerial vehicles (UAVs) in unknown environments. The existing methods often suffer from excessive computational burden, which creates the gap between theoretical approaches and practical swarm deployment. To overcome these limitations, this paper proposes a distributed cooperative planning system (DCPS). The system consists of two levels: a trajectory planning level and a cooperative trajectory planning level. On the trajectory planning level, a kinodynamic Gaussian potential B-spline (KGPB) approach is designed by combining the local kinodynamic-A-star method and the Gaussian potential field B-spline method. Specifically, in the front-end, a trajectory is first generated by using the local kinodynamic-A-star method based on kinematics-dynamics constraints. And then, in the back-end, the trajectory is further optimized by using the Gaussian potential B-spline (GPB) method. On the cooperative trajectory planning level, a back field neighbor replanning (BFNP) approach is proposed, where each UAV only needs to communicate with the nearest neighbors. According to the potential collision region of the front UAV, the trajectory of the current UAV is dynamically improved to ensure safe and stable flight such that communication costs are significantly improved. Finally, the simulation results demonstrate that the proposed DCPS achieves at least a 27% increase in average velocity and a 24% reduction in traversal time for the UAV swarm compared to prior methods. The experimental outcomes provide further validation that the proposed DCPS can generate efficient and safe trajectories. For particular cases, the UAV operates at 97% of its maximum possible velocity. The proposed DCPS provides a reliable solution for a group of unmanned aerial vehicles in unknown environments.

In this paper, a real-time Gaussian potential B-Spline planner (RGPB-Planner) is proposed to address the problem of safe flight planning for unmanned aerial vehicles (UAVs) in complex environments. Existing trajectory optimization approaches, such as B-Spline and safe-corridor methods, can generate smooth trajectories. However, these approaches often suffer from high computational cost, local collisions near obstacles, or instability in dense environments. To fill this gap, the proposed RGPB-Planner integrates a Gaussian potential with B-Spline optimization, which ensures real-time performance, smoothness, and safety by constraining control points within the safe potential region. The proposed RGPB-Planner comprises two modules: the front-end and back-end modules. In the front-end module, a local dynamic A-star graph search is used to find the shortest path for UAVs to safely reach target points in complex environments. In the back-end module, the front-end shortest path is optimized using a Gaussian potential B-Spline trajectory optimization method that considers the UAV's kinematics and dynamics constraints. Finally, simulation results demonstrate that the proposed RGPB-Planner reduces flight time by 15%-19%, increases average velocity by 13%-21%, and shortens trajectory length by 2%-3% compared with Fast-Planner and Ego-Planner, while real-world experiments validate its real-time capability and flight safety. Therefore, the proposed RGPB-Planner provides a reliable solution to ensure safe flight planning in complex environments.

This paper studies the consensus tracking control of networked stochastic leader-following multi-agent systems (MASs) with multiplicative and additive time-varying actuator failures under random communication topology switching. Considering the measurement noise generated by information transmission in networked systems, the stochastic MASs model with multiplicative noise is established. Meanwhile, the random time-varying loss of actuator effectiveness failure and bias faults are taken into account. Based on the neighbors’ and leaders’ state, the distributed adaptive fault-tolerant consensus tracking control protocols are proposed under the case of Markovian and semi-Markovian switching topology. Using stochastic system theory and Lyapunov theorem, sufficient conditions of the mean-square practical stability for leader-following consensus tracking are obtained. Results show that under the proposed distributed adaptive fault-tolerant control (DAFTC) protocols, the follower agents can track the leader under actuator constrains and random switching topology. Finally, the effectiveness of the mentioned control protocols are verified the numerical simulations.

This article deals with the problem of target tracking and detecting in unknown environments by designing two new algorithms for an autonomous aerial vehicle (AAV). First, an auto-Gaussian-GRU-predictive (AGUP) algorithm is designed to solve the tracking problem of a dynamic target in unknown environments. By integrating Gaussian process regression and gated recurrent unit neural networks, the AGUP algorithm can predict the motion trajectory of a dynamic target. Second, a Tabu search interpolated B-spline (TBL) algorithm is also proposed to solve the problem of optimal path planning for multiple stationary targets. The TBL algorithm can efficiently plan the visiting paths and also can enable the path smooth. Third, both AGUP and TBL algorithms are combined with the model predictive control (MPC) approach in order to guide AAVs to track and detect the targets. Finally, simulation and experimental results show that the AGUP-MPC algorithm exhibits excellent tracking capability. In addition, the TBL-MPC algorithm effectively plans the optimal and smooth detection path and controls AAVs to orderly visit multiple stationary targets.

In this paper, a distributed autonomous formation tracking method for unmanned aerial vehicle (UAV) swarm is proposed, aiming to improve swarm's target tracking capability in complex environments. First, the collision adjustment region is designed, which enables UAVs to fly safely and efficiently under lightweight local neighborhood communication condition, and overcomes the limitation of traditional swarm control on global information dependence. Second, an autonomous formation approach under local neighborhood communication method is proposed. The target position for each UAV is determined by offset relative to formation center in conjunction with the average of positions from all UAVs, dynamically determining target position without the requirement for a preset fixed position. This policy provides flexible formation and effectively reduces the collision risk. Finally, the above method is combined with obstacle avoidance capability, solving the problem of real-time dynamic target tracking in complex environments. Simulation results validate that UAVs can effectively track dynamic target in complex environments.

In unstructured environments, it is crucial to quickly and accurately calculate the pose of the object before using a manipulator to grip it. Irrelevant point clouds will increase computation time and decrease the probability of successful grasps. In this study, a novel pose estimation approach for manipulator grasping based on Point Pair Features (PPF) and Speeded Up Robust Features (SURF) is presented. The pose estimation procedure consists of two phases: offline and online. 3D object models are employed in the offline stage to build the model database and the ideal grasping position database. The SURF and Random Sample Consensus (RANSAC) algorithms are utilized to detect and mark the target's region in RGB images during the online stage. Segmenting the target point clouds will reduce the influence of redundant point clouds on pose estimation. The pose is calculated by a voting approach, followed by the use of iterative closest point (ICP) and pose clustering to optimize the pose. The experimental grasping results showed that the proposed method significantly improves both the success rate of grasps and real-time performance.

In this paper, a new deployable grasping mechanism for non-cooperative space debris is proposed and developed. This grasping mechanism consists of three robotic fingers connected to a platform. Each finger is developed by combining a series of scissors mechanisms, in such a way that one mechanism drives the next. A half scissors mechanism is used at the end of finger as a its tip. These fingers are deployable and their length increases and decreases with the closing and opening of the scissors mechanism. Each deployable modules is equipped with a grasp driver mechanism, which can gradually bend the finger during the process of increase in its length, in order to accomplish the grasping of the non-cooperative space debris. Each finger is designed as an under-actuated mechanism, to save the development cost and make the finger lightweight. A special mechanism is developed in the platform of the grasping mechanism, such that single motor can be used to deploy and bend all the fingers, simultaneously. In the end, the validation of the working and effectiveness of the proposed deployable grasping mechanism is given through simulations and experimental work. It can be observed through the results that the proposed mechanism is able to grasp large objects with simultaneous deployment and bending of all fingers by using single motor.

This paper investigates the fixed-time consensus (FxTC) problem for second-order multi-agent systems (MASs) with uncertain nonlinear dynamics under directed topology. In MASs communication network, bandwidth is limited, which can lead to the communication congestion issue. This issue becomes more critical in fully connected or high-density topologies. To alleviate the communication burden and sensing cost, a novel dynamic event-triggered (DET) mechanism is proposed, which accumulates state deviations over time and incorporates an inverse error term to suppress unnecessary triggering events, particularly after consensus is achieved. This design guarantees the exclusion of Zeno behavior and significantly improves communication efficiency while maintaining high accuracy in both position and velocity states. A FxTC strategy is implemented for MASs coordination, based on the constructed DET rule with a new inverse error term. Simulation results demonstrate the availability of the proposed DET-FxTC controller in achieving consensus while greatly reducing triggering frequency.

This article deals with the recovery flight problem of flapping-wing micro-aerial vehicles under extreme attitude by using a reinforcement learning approach. First, the reinforcement learning-based control policy is proposed to enable the flapping-wing micro-aerial vehicles to be recovery flight rapidly and keep the angular acceleration as small as possible. Then, a hybrid control approach is designed to significantly improve the flight stability by combining the reinforcement learning-based control approach with the proportional-derivative control approach. Finally, simulation results show the effectiveness of the reinforcement learning-based method and the hybrid control method for the flapping-wing micro-aerial vehicles under extreme attitudes.

Images captured by underwater robots often suffer from issues such as blurring and colour distortion, which hinder effective feature extraction and target recognition in underwater environments. To address these challenges, this paper proposes a novel underwater image enhancement method based on generative adversarial networks (GANs), termed multiple colour space underwater generative adversarial network (MCS-UGAN). The proposed method is built upon a GAN framework, consisting of a generator and a discriminator. The generator comprises two main modules: a deblurring module and a colour correction module. The deblurring module innovatively incorporates an efficient multi-scale feature extraction technique and an attention mechanism, which enhances object contours while preserving fine image details. The colour correction module integrates residual blocks into the U-Net architecture, effectively mitigating the problems of gradient vanishing and explosion during backpropagation in underwater image enhancement networks, thereby enhancing the network's feature learning capability. This design corrects colour distortions while preserving edge information in the image. The discriminator adopts the PatchGAN structure, which focuses on the local regions of the image, significantly improving the generator's ability to restore high-frequency details and thus enhancing the quality of the generated images. Experimental results on benchmark datasets demonstrate that, compared to existing methods, MCS-UGAN achieves superior performance in terms of peak signal-to-noise ratio, structural similarity index measure, underwater image quality measure, and underwater colour image quality evaluation, with average values of 26.24, 0.91, 3.13, and 0.64, respectively. Results from real-world applications further show that MCS-UGAN effectively increases the number of extracted corner points, validating its practicality and effectiveness. The code is available at https://github.com/invincibility6/MCS-UGAN.git

In this paper, the trajectory tracking problem of quadrotor UAV under multi-source disturbances is investigated. Considering that quadrotor UAV are often subjected to various external disturbances during flight, including wind, airflow variations, and parameter influence, a quadrotor UAV dynamic model incorporated these disturbances is established. To address the influence of multi-source disturbances on quadrotor UAV trajectory tracking, a trajectory tracking method combining Model Predictive Control (MPC) and an Extended Kalman Filter (EKF) is designed.This approach improves the precision of quadrotor UAV trajectory tracking. Finally, a simulation experiment is performed to evaluate the validity of the presented control algorithm.

In this paper, two technologies are proposed to deal with the problem of flight safty of multiple unmanned aerial vehicles (UAVs) in unknown environments. One technology is to optimize the front-end path generated by traditional path planning methods in order to better match the dynamics of UAVs to obtain the back-end movement trajectories of UAVs. The other technology is to introduce the collision detection adjustment region such that collision avoidance can be realized for multiple UAVs by dynamic replanning of UAV’s trajectory under local neighborhood communication. Finally, according to simulation and real-world experimental results, the effectiveness of the proposed technologies is verified for the flight safty of multiple UAVs in unknown environments.

With the acceleration of underground space development and infrastructure intelligence, closed environments such as tunnels and tube corridors put forward higher requirements for efficient and reliable unmanned autonomous inspection systems. Aiming at the challenges of missing GPS signals, complex structures, and unknown obstacles in unknown tunnels, this paper proposes a visual perception-based unmanned autonomous navigation and inspection system. The system adopts a three-level planning framework to complete mission scheduling and path generation, and constructs a 3D dynamic map with visual depth map, and realizes state tracking and future position prediction of dynamic obstacles by fusing Kalman filtering. On this basis, a trajectory optimization model based on cubic B-splines is constructed, and multi-objective optimization is carried out by combining the cost functions of speed, acceleration and obstacle distance. In order to enhance the dynamic obstacle avoidance capability, the fallback horizon distance field is introduced to impose time-varying obstacle avoidance cost on the area that may be occupied by future obstacles. Eventually, the system can realize autonomous navigation, environment mapping, dynamic obstacle avoidance and closed-loop mission execution for UAV in unknown tunnel environments. Simulation experiments verify the stability and practicality of this paper’s method in unknown complex tunnel environments.

In this paper, we propose a novel framework which can minimize event-triggered frequency while maintaining consensus control performance in multi-agent systems. Distributed event-triggering condition is learned automatically through re-inforcement learning techniques, integrating with existing consensus controllers. Different from conventional event-triggered consensus control methods that rely on Lyapunov-based stability analysis, the proposed data-driven framework reduces design conservatism through adaptive learning mechanisms. Experimental results demonstrate the effectiveness of our method in achieving substantial reductions in communication cost while preserving system stability and consensus performance.