Most simultaneous localization and mapping (SLAM) systems perform well in texture-rich environments but suffer from significant rotational drift in texture-limited scenes. Moreover, real-time performance, accuracy, and efficiency in map segmentation remain challenging. To address these problems, we propose a novel real-time RGB-D SLAM system based on the Manhattan World (MW) assumption. This system detects Manhattan frames (MF) based on lines and planes, and performs pose estimation based on the Manhattan coordinate system in MF. For non-MF, constraints are added to minimize reprojection errors for parallel and perpendicular lines and planes. The proposed method is well-suited for texture-limited environments and can effectively reduce the long-term trajectory drift caused by rotation estimation. For mapping and segmentation, we propose an improved real-time incremental plane-based segmentation method, which determines segmentation boundaries using geometric cues from depth images and subsequently refines label association through pairwise plane-label confidence. The accuracy of the proposed method is evaluated both on public datasets and in real-world scenarios, which exceeds that of the state-of-the-art methods.

In this article, we investigate the cooperative robust parallel operation problem of a general linear uncertain system driven by multiple actuators. Compared with the existing work, the uncertainties for both the controlled system and the actuator systems, as well as the disturbances for actuator systems, are all taken into consideration. To solve the problem, we develop a dynamic output feedback controller based on the internal model principle. First, a general framework for solving the cooperative robust parallel operation problem is established. By establishing some technical lemmas, we reveal the relationship between the system input sharing of actuators and the robust output regulation property. Then, for different network conditions, we propose the gain matrix design methods in terms of the network connectivity property and the system matrix of the dynamic controller. It guarantees the solvability of the regulator equations and the stability property of the system matrix associated with the closed-loop system. Resorting to the internal model principle, we show that the cooperative robust parallel operation problem can be solved in spite of system uncertainties and disturbances. Finally, the effectiveness of the proposed distributed internal model approach is verified by its application to a shaft driven control with fifteen electric actuator motors.

This paper investigates the distributed Nash equilibrium seeking problem of $N$ players with discrete-time dynamics under jointly strongly connected switching networks. First, by integrating gradient play technique with a novel discrete-time distributed observer design, we develop a discretetime distributed Nash equilibrium seeking controller to solve the problem. Next, we establish two technical lemmas to analyze the uniform exponential stability of discrete-time linear switched systems. Then, we obtain the exponential stability of the interconnected closed-loop systems by constructing a Lyapunov function and providing a controller gain design method with the small gain arguments. Finally, the discrete-time distributed Nash equilibrium seeking is achieved over jointly strongly connected switching communication networks. The effectiveness of the developed approach is demonstrated by a numerical simulation example.

Autonomous aerial swarms demonstrate significant potential for a range of applications, such as environmental monitoring, disaster response, and search-and-rescue operations. However, achieving safe and decentralized navigation in dynamic, cluttered environments remains a fundamental challenge, particularly under strict constraints of onboard sensing and computation. Classical modular pipelines suffer from latency accumulation and limited scalability, while fully end-to-end Reinforcement Learning (RL) approaches often face severe sim-to-real degradation and lack safety or stability guarantees. To address these challenges, this paper proposes a novel learning-based decentralized navigation framework that integrates a LiDAR-based RL policy with a Safety-assured Nonlinear Model Predictive Controller (SA-NMPC) for reliable execution. The proposed framework features a biologically-inspired decoupled hierarchical architecture: the RL front-end generates agile, short-horizon navigation commands based on raw Light Detection and Ranging (LiDAR) scans, while the SA-NMPC back-end ensures dynamically feasible tracking and active disturbance rejection. To ensure safe operation in dynamic scenes, an asynchronous dual-stream perception system is employed to enhance the capabilities of dynamic obstacle tracking and static map maintenance. The proposed framework has been validated through extensive simulation and real-world experiments, including the 2025 IEEE IROS Aerial Autonomy Challenge and multi-quadrotor swarm flights. The system demonstrates zero-shot sim-to-real transfer capability, robust performance in dynamic environments, and significant improvements over both classical and learning-based baselines.

Wearable and autonomous sensing systems require efficient energy conversion under low-frequency, bidirectional mechanical excitations generated by daily human motion. This work presents a novel dual-clutch ratchet energy harvester (DREH) that converts bidirectional heel motions into unidirectional high-speed rotation, enabling direct DC power generation without conventional AC-DC rectification losses. The dual-clutch architecture ensures robust motion rectification under weak excitation and enables recovery of elastic potential energy during the liftoff phase, thereby improving effective energy utilization. A coupled dynamic-electromechanical model is developed to analyze the system behavior and predict electrical output. A compact prototype (86.4 cm(3), 89 g) is fabricated and experimentally evaluated under pseudowalk and natural walking conditions. Experimental results show that elastic energy recovery increases output power by 59.5% at 0.3 Hz; at 6 km/h with a 75 Omega load, the prototype delivers 82 mW peak and 28 mW average power with a driving force below 14 N. System-level validation demonstrates continuous operation of a wireless sensing node, confirming the DREH as an effective self-sustained power source for wearable and industrial sensing applications.

Capturing aerial intruders with multiple uncrewed aerial vehicles (UAVs) using a net is challenging, as it requires safe target tracking without damaging the net. This article presents a coordinated controller based on Markov random field (MRF) with identified parameters for safe multi-UAV net capture. First, energy functions derived from UAV dynamic states form the MRF-based control structure, enabling coordinated flight without assuming a fixed net shape. Next, a reference velocity is designed to track the intruder, and the control input is obtained by solving the MRF via mean-field approximation. Finally, Gaussian process regression (GPR) identifies the MRF energy function parameters to regulate inter-UAV distances and improve capture safety. An optimized training point selection strategy improves GPR parameter estimation accuracy. Simulations and real-world experiments, including comparisons with existing net capture strategies, confirm the effectiveness of the proposed controller and the GPR-based parameter identification with the training point selection strategy.

In this article, we investigate the distributed optimization problem of heterogeneous general linear multiagent systems by the adaptive dynamic programming (ADP) approach over directed communication networks. A distinctive feature of this work is the development of a data-driven approach that eliminates the need for prior knowledge of system dynamics for all agents. To address the challenges posed by unknown system dynamics, we utilize the ADP-based data-driven approach to develop the distributed optimization control law. First, the feedback gain of the control law is determined based on the state and input data of the controlled systems. Next, the system dynamics are reconstructed using the solved feedback gain and the running data of the controlled systems. Then, the remaining parameters in the control law are designed by solving a series of steady-state equations. Under standard assumptions and through the application of the certainty equivalence principle, we prove that the proposed approach solves the distributed optimization problem, ensuring output consensus of all agents at the optimal solution of the global cost function. Finally, the viability of our proposed approach is demonstrated through its application to optimal output power sharing control of hydraulic turbine systems and their large-scale form.

In this paper, we address the distributed Nash equilibrium seeking problem for aggregative games of N players subject to unknown disturbances over strongly connected networks. Compared with existing works, the general linear dynamics, general directed and strongly connected networks, as well as unknown disturbances are tackled simultaneously in the aggregative games. First, by introducing certain coordinate transformation and feedback linearization method, we develop a distributed gradient-based Nash equilibrium seeking law. A dynamic average consensus dynamics is designed to deal with the challenge by unbalance of general strongly connected networks. By the graph-related property and converse Lyapunov theorem, we establish the global exponential stability of a linear system and a class of nonlinear systems, respectively. Then, we propose a gain design method to obtain the stability of the nonlinear closed-loop system, which is not in the lower triangular form. Inspired by the output regulation theory, we design an internal model and an adaptive dynamics to tackle the unknown disturbances. Resorting to the perturbation theory and the internal model principle, we demonstrate that distributed Nash equilibrium seeking for aggregative games of N players with general linear systems subject to unknown disturbances over strongly connected networks can be achieved. Finally, the effectiveness of the proposed distributed Nash equilibrium seeking approaches are verified by their applications to some simulation examples. (c) 2025 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

This paper investigates the distributed estimation problem for discrete-time uncertain linear time invariant systems under the joint observability condition. The problem is highly challenging and remained unresolved until it was recently addressed by an adaptive approach. This papersolvesthisproblemusinganonadaptive approach that introduces new features. An observability transformation is first introduced to decompose the uncertain system, thereby facilitating the parameter and state reconstruction by the two nonlinear mappings. Then, a novel class of distributed nonadaptive observers is pro posed, composed of a nonlinear distributed Luenberger type observer dynamics and two nonlinear mappings. By Lyapunov stability analysis and nonlinear mappings recon struction, we show that the parameter and state estimation are achieved asymptotically for discrete-time jointly observable uncertain linear systems. The proposed nonadaptive scheme enriches design tools for robust control, and circumvents the convergence and stability issues inherent in adaptive approaches as well. Besides, the observer is fully distributed, requiring only local measurements and neighbor-to-neighbor communication, which ensures computational efficiency and scalability.

Achieving robust agile quadrotor flight under unknown external disturbances and parametric uncertainty remains a significant challenge. Inspired by robust output regulation theory, this paper proposes ${\mathcal {I} \mathcal {M}}$-NMPO, an ${\mathcal {I}}$nternal ${\mathcal {M}}$odel-based Nonlinear Model Predictive Optimization framework for robust optimal planning and control during agile flight. The core insight is the systematic decoupling of uncertain dynamics from the optimization loop via the internal model principle (IMP), enabling real-time disturbance learning and rejection while reducing reliance on high-fidelity physical models. This allows both time-optimal planning and nonlinear model predictive control (NMPC) to operate on the same disturbance-decoupled nominal dynamics. The proposed framework comprises three main components: translational and rotational nonlinear internal model compensators for real-time disturbance rejection, an NMPC optimization stabilizer for constrained agile trajectory tracking, and a receding-horizon polynomial waypoint allocation module for efficient online time-optimal reference generation. Robust constraint satisfaction, recursive feasibility and asymptotic stability are guaranteed through rigorous theoretical analysis. The effectiveness and generalizability are validated through extensive flight experiments under various disturbances, including unknown payloads, persistent fan-induced winds, and time-varying gusts, across quadrotors with different wheelbases. The framework was successfully deployed in the 2024 DJI Robomaster Intelligent MAV Championship for Planning and Control of Quadrotors, where it completed challenging racing courses at speeds up to 23.3 m/s, ranking first and finishing in less than one-third of the time taken by the second-place.

The classical linear quadratic regulation (LQR) problem of linear systems by state feedback has been widely addressed. However, the LQR problem by dynamic output feedback with optimal transient performance remains open. The main reason is that the observer error inevitably leads to suboptimal transient performance of the closed-loop system. In this article, we propose an optimal dynamic output feedback learning control approach to solve the LQR problem of linear continuous-time systems with unknown dynamics. In particular, we propose a novel internal dynamics called the internal model. Unlike the classical $p$-copy internal model, it is driven by the input and output of the system, and the role of the proposed internal model is to compensate for the transient error of the observer such that the output feedback LQR problem is solved with guaranteed optimality. A model-free learning algorithm is developed to estimate the optimal control gain of the dynamic output feedback controller. The algorithm does not require any prior knowledge of the system matrices or the system's initial state, thus leading to an optimal solution to the model-free LQR problem. The effectiveness of the proposed method is illustrated using an aircraft control system.

This paper proposes a constructive distributed adaptive observer design approach for jointly observable discrete-time uncertain linear time-invariant (LTI) systems over time-varying communication networks. In comparison with existing works, the approach developed in this work can guarantee distributed state estimation in the presence of sparse sensor arrangement, modeling uncertainties, and unreliable network communication. A discrete-time linear system decomposition method is first developed to mitigate the impact of the unknown parameters. A fully distributed discrete- time adaptive nonlinear observer is then designed for the decomposed system. The observer is composed of two time-varying discrete-time dynamics and two nonlinear mappings that establish the connection between the observed system's state and parameters and the states of the two time-varying dynamics. By establishing a discrete-time parametric representation of the measurement output, the parameter estimation problem is converted to a parameter identification problem, which is solved by the gradient-descent adaptive law in the proposed observer dynamics. The analysis shows that the estimation error system is asymptotically stable. Thus, the distributed discrete-time adaptive observer holds under the jointly observable condition in spite of system uncertainties and jointly connected communication networks. The robustness properties of the proposed distributed discrete- time adaptive observer in the presence of measurement noise are established using an input-to-state stability analysis. (c) 2024 Published by Elsevier Ltd.

This article presents a novel design of a distributed adaptive observer for distributed state estimation of continuous-time uncertain linear time-invariant systems over directed networks. In contrast to existing works, the observed system is subject to uncertainties and possibly jointly observable. The distributed estimation of such systems allows practical applications in challenging problems subject to sparse arrangement of sensors and model uncertainties. A class of fully distributed nonlinear adaptive observers is proposed to address these challenges. In particular, we introduce an observability decomposition method to decompose both the state and the unknown parameters of the observed system into an observable and an unobservable component. This decomposition circumvents the impact of the unknown parameters on existing observability decomposition methods. Two nonlinear mappings are designed to achieve the reconstruction of the system state and the unknown system parameters. A parametric representation of the output estimation error is established to convert the unknown parameter estimation problem of the observable subsystem into an unknown parameter identification problem using a linear regression equation. Using a Lyapunov stability analysis, it is shown that the system parameter can be recovered by the nonlinear mappings, while the distributed state estimation problem is solved.

In this article, we address the robust distributed Nash equilibrium seeking problem of N-player games under switching networks and communication delays. The salient feature of this work is that the switching communication networks can be uniformly strongly connected, and the communication delays are allowed to be arbitrarily unknown, time-varying and bounded. To solve the problem, we construct a distributed estimator for each player to estimate all players' strategies through unreliable communication networks. Based on the gradient play technique, we design a distributed Nash equilibrium seeking law. Then, we obtain the closed-loop system, which is an interconnected system of a nonlinear subsystem and a linear time-delay subsystem. By constructing the Lyapunov-Krasovskii functional, and designing the controller parameter in the sense of the small gain theorem, we achieve robust Nash equilibrium seeking asymptotically in spite of unreliable communication networks. Finally, we illustrate our proposed approach by its application to practical motion control of mobile robots with an experiment.

In this paper, we address the cooperative robust parallel operation problem of an electric drive shaft system. In contrast to prior research, this work explicitly incorporates system uncertainties and external disturbances affecting both the shaft and the motors, enhancing the robustness and practicality of the proposed approach. To address this challenge, we first establish a dynamic output feedback controller utilizing the internal model principle. Then, we demonstrate that the cooperative robust parallel operation of the electric drive shaft system can be achieved under directed communication networks, effectively overcoming the adverse effects of system uncertainties and external disturbances. Finally, the efficacy of our proposed distributed controller is rigorously validated through its application to an electric drive shaft system equipped with five actuator motors.

In this paper, we investigate the semi-global robust output regulation problem of a class of nonlinear networked control systems. By the emulation approach, we propose a class of sampled-data output feedback control laws to solve this problem. In particular, we first develop a general sampled-data dynamic output feedback control law and characterize the closed-loop system by a hybrid system. Then, we design the internal model based on the sampled error output of the system. Based on the internal model principle, we convert the semi-global robust output regulation problem into a semi-global robust stabilization problem of an augmented hybrid system composed of the internal model and the original system. By proposing the sampled error output feedback control law and by means of Lyapunov analysis, we obtain the maximum allowable transmission interval for sampling and show that semi-global robust stabilization of the augmented hybrid system can be achieved by the proposed sampled-data control law and thus leading to the solution of the semi-global robust output regulation problem. Finally, we apply the proposed control approach to two practical applications to verify the effectiveness of the proposed control approach.

This paper investigates the multi-player non-zero-sum game problem for unknown linear continuous-time systems with unmeasurable states. By only accessing the data information of input and output, a data-driven learning control approach is proposed to estimate N-tuple dynamic output feedback control policies which can form Nash equilibrium solution to the multi-player non-zero-sum game problem. In particular, the explicit form of dynamic output feedback Nash strategy is constructed by embedding the internal dynamics and solving coupled algebraic Riccati equations. The coupled policy-iteration based iterative learning equations are established to estimate the N-tuple feedback control gains without prior knowledge of system matrices. Finally, an example is used to illustrate the effectiveness of the proposed approach.

We present an air-to-air multi-sensor and multi-view fixed-wing UAV dataset, MMFW-UAV, in this work. MMFW-UAV contains a total of 147,417 fixed-wing UAVs images captured by multiple types of sensors (zoom, wide-angle, and thermal imaging sensors), displaying the flight status of fixed-wing UAVs of different sizes, appearances, structures, and stabilized flight velocities from multiple aerial perspectives (top-down, horizontal, and bottom-up views), aiming to cover the full-range of perspectives with multi-modal image data. Quality control processes of semi-automatic annotation, manual check, and secondary refinement are performed on each image. To the best of our knowledge, MMFW-UAV is the first one-to-one multi-modal image dataset for fixed-wing UAVs with high-quality annotations. Several mainstream deep learning-based object detection architectures are evaluated on MMFW-UAV and the experimental results demonstrate that MMFW-UAV can be utilized for fixed-wing UAV identification, detection, and monitoring. We believe that MMFW-UAV will contribute to various fixed-wing UAVs-based research and applications.

In this article, we investigate cooperative output regulation of heterogeneous linear multiagent systems subject to an uncertain exosystem over directed communication networks. Compared with existing results, a fully distributed output-based adaptive observer is developed that eliminates the requirement for global information of communication networks and improves practical feasibility. By establishing a sequence of technical lemmas, it is demonstrated that the proposed distributed adaptive observer attains distributed exponential estimation of the unknown parameter and the state of the uncertain exosystem under directed communication networks. Leveraging the certainty equivalence principle, both the distributed state feedback control law and the distributed measurement output feedback control law are proposed to solve the problem. Moreover, two applications, active automotive suspension control under unknown road profiles and cooperative output regulation of multiple LCL-coupled inverter-based distributed generation systems, are provided to verify the proposed approach.

3D human pose estimation (3DHPE) based on a single RGB image is a fundamental problem in the field of computer vision. The estimation accuracy and speed of 3DHPE directly affect the practical applicability. However, existing methods often improve accuracy by using complex network architectures or multi-stage processing. These approaches result in more computational overhead and slower estimation speed. To balance estimation accuracy and speed, we need a more efficient approach. Motivated by the need for both estimation accuracy and speed, we propose the I-KDnet, an end-toend model, that achieves precise 3D pose estimation at a high speed. Specifically, we design an Idealized Knowledge Distillation (I-KD) training approach, an idealized variant of online knowledge distillation. During training, I-KD enhances the training process, similar to online knowledge distillation. During inference, it does not introduce any additional computational overhead. Additionally, compared to online knowledge distillation, the I-KD approach is easier to implement and more effective. Based on this approach, I-KDNet achieves the best accuracy on the Human3.6M benchmark with high estimation speed.