Learning from demonstration(LfD) allows for the effective transfer of human manipulation skills to a robot by building a model that represents these skills based on a limited number of demonstrated ***,a skilllearning...
详细信息
Learning from demonstration(LfD) allows for the effective transfer of human manipulation skills to a robot by building a model that represents these skills based on a limited number of demonstrated ***,a skilllearning model that can comprehensively satisfy multiple requirements,such as computational complexity,modeling accuracy,trajectory smoothness,and robustness,is still ***,this work aims to provide such a model by employing fuzzy ***,we introduce an LfD model named Takagi-Sugeno-Kang fuzzy system-based movement primitives(TSKFMPs),which exploits the advantages of the fuzzy theory for effective robotic imitation learning of human *** work formulates the TSK fuzzy system and gradient descent(GD) as imitation learning models,leveraging recent advancements in GD-based optimization for fuzzy *** study takes a two-step strategy.(ⅰ) The input-output relationships of the model are established using TSK fuzzy systems based on demonstration *** this way,the skill is encoded by the model parameter in the latent space.(ⅱ) GD is used to optimize the model parameter to increase the modeling accuracy and trajectory *** further explain how learned trajectories are adapted to new task scenarios through local *** conduct multiple tests using an open dataset to validate our method,and the results demonstrate performance comparable with those of other ***,we implement it in a real-world case study.
This article addresses the model-and data-based event-triggered consensus of heterogeneous leader/follower multi-agent systems(MASs). A dynamic periodic transmission protocol is developed to alleviate the communicatio...
详细信息
This article addresses the model-and data-based event-triggered consensus of heterogeneous leader/follower multi-agent systems(MASs). A dynamic periodic transmission protocol is developed to alleviate the communication and computational burden, where the followers can interact locally with neighbors to approach the dynamics of the leader. Capitalizing on a discrete-time looped-functional, a model-based consensus condition for the closed-loop MASs is derived as linear matrix inequalities(LMIs), along with a design method for obtaining distributed event-triggered controllers and the associated triggering *** collecting noise-corrupted state-input measurements in offline open-loop experiments, a data-based leader/follower MAS representation is derived and employed to address the data-driven consensus control problem without explicit MAS models. This result is subsequently generalized to guarantee an H∞-consensus control performance. Finally, a simulation example is given to corroborate the efficiency of the proposed distributed triggering scheme and the data-driven consensus controller.
Recently, a reference derived some new higher-order output tracking properties for direct model reference adaptive control（MRAC） of linear time-invariant（LTI） systems: limt→∞ e（i）（t） = 0, i = 1,..., n*-1, wh...
详细信息
Recently, a reference derived some new higher-order output tracking properties for direct model reference adaptive control（MRAC） of linear time-invariant（LTI） systems: limt→∞ e（i）（t） = 0, i = 1,..., n*-1, where n*and e（i）（t） denote the relative degree of the system and the i-th derivative of the output tracking error, respectively. However, a naturally arising question involves whether indirect adaptive control（including indirect MRAC and indirect adaptive pole placement control） of LTI systems still has higher-order tracking properties. Such properties have not been reported in the literature. Therefore, this paper provides an affirmative answer to this question. Such higher-order tracking properties are new discoveries since they hold without any additional design conditions and, in particular, without the persistent excitation condition. Given the higher-order properties, a new adaptive controlsystem is developed with stronger tracking features.（1） It can track a reference signal with any order derivatives being unknown.（2） It has higher-order exponential or practical output tracking properties.（3） Finally, it is different from the usual MRAC system, whose reference signal's derivatives up to the n*order are assumed to be known. Finally, two simulation examples are provided to verify the theoretical results obtained in this paper.
This paper presents a novel approach for addressing the finite-time distributed formation maneuvering (FTDFM) of multiple unmanned surface vehicles (USVs), which takes into account the challenges posed by velocity and...
详细信息
This paper presents a novel approach for addressing the finite-time distributed formation maneuvering (FTDFM) of multiple unmanned surface vehicles (USVs), which takes into account the challenges posed by velocity and error constraints. Each USV is subject to parameter uncertainty, ocean disturbance, actuator fault, and input saturation, making the task of achieving reliable and accurate formation particularly challenging. To overcome these challenges and meet practical requirements, a finite-time (FT) performance function is selected as the constraint function, which ensures that the velocity and error of each USV stay within a given bounded set within a known time. Using FT stability theory, a new framework is proposed that integrates a tangent-type barrier function and an improved backstepping approach to handle uncertainties and constraints. In this approach, a tracking differentiator (TD) is introduced to replace the virtual controller's derivative, and a smooth function is used to address the input saturation, effectively reducing the complexity and dynamic order of the algorithm. The proposed controller is capable of ensuring the realization of the desired formation within a finite time while maintaining the constraints without deviation. Additionally, by using the auxiliary variable technique, the proposed control method can also be applied to USVs with underactuated models. Simulation examples are provided to demonstrate the efficacy of the proposed control algorithm in achieving accurate and reliable formation maneuvering of multiple USVs under various constraints. IEEE
Recent years have seen a rising interest in distributed optimization problems because of their widespread applications in power grids, multi-robot control, and regression *** the last few decades, many distributed alg...
Recent years have seen a rising interest in distributed optimization problems because of their widespread applications in power grids, multi-robot control, and regression *** the last few decades, many distributed algorithms have been developed for tackling distributed optimization problems. In these algorithms, agents over the network only have access to their own local functions and exchange information with their neighbors.
作者：
Liu, QimingCui, XinruLiu, ZheWang, HeshengDepartment of Automation
Shanghai Jiao Tong University Shanghai China MoE Key Lab of Artificial Intelligence
AI Institute Shanghai Jiao Tong University Shanghai China Department of Automation
Key Laboratory of System Control and Information Processing of Ministry of Education Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education Shanghai Engineering Research Center of Intelligent Control and Management Shanghai Jiao Tong University Shanghai China
Target search in unknown environments places high demands not only on an autonomous vehicle's ability to perceive and interpret target cues, but also on its conscious of collecting these cues by active exploration...
详细信息
Target search in unknown environments places high demands not only on an autonomous vehicle's ability to perceive and interpret target cues, but also on its conscious of collecting these cues by active exploration. While existing navigation methods have successfully built target-driven policies by maintaining memory of explored areas, there has been a lack of focus on facilitating target-aware exploration-the informative frontier information at unexplored yet visible areas is often overlooked. In this paper, we introduce a novel topology-based memory structure, Frontier-enhanced Topological Memory (FTM), and a Hierarchical Topology Encoding and Extraction (HTEE) module, fostering the autonomous vehicle's awareness of both environmental exploration and target approach. Specifically, FTM innovatively incorporates informative ghost nodes on traditional topological map to represent unexplored yet visible regions. We leverage an online-trained implicit scene representation to estimate the positions and generate features of these ghost nodes. The HTEE then employs implicit graph convolutions and attention mechanisms to extract cognitive information from FTM, taking into account the hierarchical memory structure, target cues, and current state. Our design bolsters cognitive navigation decisions. The experiments in the high-fidelity environments, including performance tests, visualizations, and interpretability experiments, validate the effectiveness of our approach in enhancing the vehicle's exploratory behavior. The improved exploration awareness for target cue collection, in turn, enhances the success rate and path efficiency of target search. Furthermore, we demonstrate the adaptability of our algorithm in real-world physical environments. IEEE
Ocean engineering demands systemic solutions for mission-critical demands and technological challenges. From an industrial control perspective, development engineers need to factor in both costs and established standa...
详细信息
Aircraft final assembly line(AFAL)involves thousands of processes that must be completed before ***,the heavy reliance on manual labor in most assembly processes affects the quality and prolongs the delivery *** the a...
详细信息
Aircraft final assembly line(AFAL)involves thousands of processes that must be completed before ***,the heavy reliance on manual labor in most assembly processes affects the quality and prolongs the delivery *** the advent of artificial intelligence of things(AIoT)technologies has introduced advancements in certain AFAL scenarios,systematically enhancing the intelligence level of the AFAL and promoting the widespread deployment of artificial intelligence(AI)technologies remain significant *** address these challenges,we propose the intelligent and collaborative aircraft assembly(ICAA)framework,which integrates AI technologies within a cloud-edge-terminal *** ICAA framework is designed to support AI-enabled applications in the AFAL,with the goal of improving assembly efficiency at both individual and multiple process *** analyze specific demands across various assembly scenarios and introduce corresponding AI technologies to meet these *** three-tier ICAA framework consists of the assembly field,edge data platform,and assembly cloud platform,facilitating the collection of heterogeneous terminal data and the deployment of AI *** framework enhances assembly efficiency by reducing reliance on manual labor for individual processes and fostering collaboration across multiple *** provide detailed descriptions of how AI functions at each level of the ***,we apply the ICAA framework to a real AFAL,focusing explicitly on the flight controlsystem testing *** practical implementation demonstrates the effectiveness of the framework in improving assembly efficiency and promoting the adoption of AIoT technologies.
Timely transmission line fire inspections are vital for power system safety. Although deep learning models are widely used for flame detection, struggle with small target recognition due to background interference and...
详细信息
This paper investigates an interval analysis method for neural networks and applies it to fault detection for systems with unknown but bounded measurement noise. First, a novel interval analysis method is presented, w...
详细信息
This paper investigates an interval analysis method for neural networks and applies it to fault detection for systems with unknown but bounded measurement noise. First, a novel interval analysis method is presented, which can compute the bounds of the output of a feedforward neural network subject to a bounded input. By applying the proposed interval analysis method to a network trained with fault-free system data, adaptive thresholds for fault detection are computed. Finally, one can acquire fault detection results via a fault detection strategy. The proposed method can achieve tight bounds of the network output and employ simple operations, which leads to accurate fault detection results and a low computational burden.A numerical simulation and an experiment on an AC servo motor are given to illustrate the effectiveness and superiority of the proposed method.
暂无评论