This paper investigates the consensus problem for a set of nonlinear multi-agent systems with nonlinear interconnections. First, in order to reduce the communication burden in the multi-agent network, a distributed ev...
详细信息
This paper investigates the consensus problem for a set of nonlinear multi-agent systems with nonlinear interconnections. First, in order to reduce the communication burden in the multi-agent network, a distributed event-triggered consensus control is designed by taking into account the effect of the nonlinear interconnections. Then, based on the Lyapunov functional method and the Kronecker product technique, sufficient conditions are obtained to guarantee the consensus in the form of linear matrix inequality (LMI). Finally, a simulation example is proposed to illustrate the effectiveness of the developed theory.
Neuroplasticity has been demonstrated to play an important role in function recovery *** this paper,stroke patients and controls were subjected to functional magnetic resonance imaging(fMRI) study for *** group indepe...
详细信息
Neuroplasticity has been demonstrated to play an important role in function recovery *** this paper,stroke patients and controls were subjected to functional magnetic resonance imaging(fMRI) study for *** group independent componentanalysis was used to get the time courses of the interested regions in *** investigate the reorganization of cerebral cortex after stroke,Structural Equation Modeling(SEM) was applied to construct a locomotor brain network *** results were analyzed to investigate thebrain activity changes involved in the cerebral motor cortex circuitry aroused by the hand *** contrast to healthy people,we found that the brain activity changes of the stroke patientscould not onlyshow the local changes around injured regions,but also the global transformation of brain related to *** the differences were examined in terms of changes in path coefficients between brain regions.
In today’s ever-changing world, the ability of machine learning models to continually learn new data without forgetting previous knowledge is of utmost importance. However, in the scenario of few-shot class-increment...
In today’s ever-changing world, the ability of machine learning models to continually learn new data without forgetting previous knowledge is of utmost importance. However, in the scenario of few-shot class-incremental learning (FSCIL), where models have limited access to new instances, this task becomes even more challenging. Current methods use prototypes as a replacement for classifiers, where the cosine similarity of instances to these prototypes is used for prediction. However, we have identified that the embedding space created by using the relu activation function is incomplete and crowded for future classes. To address this issue, we propose the Expanding Hyperspherical Space (EHS) method for FSCIL. In EHS, we utilize an odd-symmetric activation function to ensure the completeness and symmetry of embedding space. Additionally, we specify a region for base classes and reserve space for unseen future classes, which increases the distance between class distributions. Pseudo instances are also used to enable the model to anticipate possible upcoming samples. During inference, we provide rectification to the confidence to prevent bias towards base classes. We conducted experiments on benchmark datasets such as CIFAR100 and miniimageNet, which demonstrate that our proposed method achieves state-of-the-art performance.
In this paper, an improved formulation of optimal guidance law (OGL) based on genetic algorithms (GAs) is proposed. Linear quadratic optimal control theory is derived to consider terminal velocity maximisation, also G...
详细信息
Mating restriction strategies are capable of restricting the distribution of parent solutions for effective offspring generation in evolutionary algorithms (EAs). Studies have shown the importance of these strategies ...
详细信息
Sensing 3D objects is critical when 2D object recognition is not accessible. A robot pre-trained on a large point-cloud dataset will encounter unseen classes of 3D objects after deploying it. Therefore, the robot shou...
Sensing 3D objects is critical when 2D object recognition is not accessible. A robot pre-trained on a large point-cloud dataset will encounter unseen classes of 3D objects after deploying it. Therefore, the robot should be able to learn continuously in real-world scenarios. Few-shot class-incremental learning (FSCIL) requires the model to learn from few-shot new examples continually and not forget past classes. However, there is an implicit but strong assumption in the FSCIL that the distribution of the base and incremental classes is the same. In this paper, we focus on cross-domain FSCIL for point-cloud recognition. We decompose the catastrophic forgetting into base class forgetting and incremental class forgetting and alleviate them separately. We utilize the base model to discriminate base samples and new samples by treating base samples as in-distribution samples, and new objects as out-of-distribution samples. We retain the base model to avoid catastrophic forgetting of base classes and train an extra domain-specific module for all new samples to adapt to new classes. At inference, we first discriminate whether the sample belongs to the base class or the new class. Once classified at the model level, test samples are then passed to the corresponding model for class-level classification. To better mitigate the forgetting of new classes, we adopt the soft label and hard label replay together. Extensive experiments on synthetic-to-real incremental 3D datasets show that our proposed method can balance the performance between the base and new objects and outperforms the previous state-of-the-art methods.
作者:
Xin-Wu LIANGXin-Han HUANGMin WANGDepartment of Automation
Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing of Ministry of Education of China Shanghai 200240 China Department of Control Science and Engineering
Huazhong University of Science and Technology Key Laboratory of Image Processing and Intelligent Control of Ministry of Education of China Wuhan 430074 China
image-based visual servoing can be used to efficiently control the motion of robot manipulators. When the initial and the desired configurations are distant, however, as pointed out by many researchers, such a control...
详细信息
image-based visual servoing can be used to efficiently control the motion of robot manipulators. When the initial and the desired configurations are distant, however, as pointed out by many researchers, such a control approach can suffer from the convergence and stability problems due to its local properties. By specifying adequate image feature trajectories to be followed in the image, we can take advantage of the local convergence and stability of image-based visual servoing to avoid these problems. Hence, path planning in the image space has been an active research topic in robotics in recent years. However, almost all of the related results are established for the case of camera-in-hand configuration. In this paper, we propose an uncalibrated visual path planning algorithm for the case of fixed-camera configuration. This algorithm computes the trajectories of image features directly in the projective space such that they are compatible with rigid body motion. By decomposing the projective representations of the rotation and the translation into their respective canonical forms, we can easily interpolate their paths in the projective space. Then, the trajectories of image features in the image plane can be generated via projective paths. In this way, the knowledge of feature point structures and camera intrinsic parameters are not required. To validate the feasibility and performance of the proposed algorithm, simulation results based on the puma560 robot manipulator are given in this paper.
The focus of this paper is the design and development of an automatic system for microassembly. The automatic processing is made possible by (i) the development of a machine vision algorithm to identify the targets an...
详细信息
The focus of this paper is the design and development of an automatic system for microassembly. The automatic processing is made possible by (i) the development of a machine vision algorithm to identify the targets and end-effectors, (ii) an uncalibrated visual servoing algorithm to lead the end-effector to grasp the micro-pieces and then assemble the target. The experimental results demonstrate that this prototype microassembly system is effective and practicable for automatic microassembly applications.
Platoon control of autonomous industrial vehicles contributes to improving the safety and reliability of cargo transportation in complex industrial scenarios, increasing traffic efficiency and saving energy. The plato...
详细信息
As a sociological phenomenon, rumor spreading has been widely researched by sociologists and other fields’ scholars. How do people generate those strange thinking about the rumor? This paper, from artificial intellig...
详细信息
暂无评论