Genetic algorithm (GA) is an effective method for path planning problems. As a powerful variant of GA, island genetic algorithm (IGA) has considerable improvement in performance. In this paper, a new island model of G...
详细信息
The safety protection of process control systems plays a crucial role in the overall safety of critical *** have increased the complexity of existing safety protection analysis. Traditional safety analysis methods fal...
详细信息
The safety protection of process control systems plays a crucial role in the overall safety of critical *** have increased the complexity of existing safety protection analysis. Traditional safety analysis methods fall short in accounting for cyberattack factors, making it challenging to conduct safety protection analysis under cyberattacks. To address this issue, this paper presents a new safety protection analysis method that considers multiple safety factors explicitly including cyberattacks using formal verification. The method consists of three main components: exhaustive system safety specifications,formal models, and system safety validation. The system safety specification component adds a cyberattack factor to system safety requirements based on the system theory process analysis(STPA) method. The formal model component considers the system's dynamic operation process, and safety protection behaviors under typical attack behaviors. The system safety validation component validates the effectiveness of system safety protection under cyberattacks by the UPPAAL tool, from the perspective of whether system safety constraints are triggered and whether the change curve of process variables is compliant. Finally, the effectiveness of the presented approach is carried out for a simplified fluid catalytic cracking(FCC) fractionating system.
This study addresses the complexities of orchestrating multi-target transportation tasks within multi-agent systems, constrained by load capacity. The primary objective is to engineer an advanced path planning framewo...
详细信息
ISBN:
(数字)9798350356618
ISBN:
(纸本)9798350356625
This study addresses the complexities of orchestrating multi-target transportation tasks within multi-agent systems, constrained by load capacity. The primary objective is to engineer an advanced path planning framework that assures collision avoidance and optimal load distribution. Diverging from conventional single-path strategies, this research incorporates load factors and requires continuous dynamic multi-target point and path planning to ensure agents can efficiently navigate through a series of predetermined key points. To address this complex requirement, a two-stage multi-agent task allocation and path planning method is proposed. First, an initial solution is obtained using a suboptimal algorithm, followed by optimization iterations using the large neighborhood search algorithm to improve task allocation. In the second stage, an accelerated algorithm based on priority search is used to plan optimal paths for each agent in a predetermined order. The proposed algorithm's effectiveness is comprehensively evaluated through a series of experimental evaluations and comparisons with the commercial solver Gurobi within a limited time. The results show that the proposed method achieves optimality in both running time and minimum path cost while ensuring the load balance of agents.
The tracking of maneuvering targets in radar networking scenarios is studied in this *** the interacting multiple model algorithm and the expected-mode augmentation algorithm,the fixed base model set leads to a mismat...
详细信息
The tracking of maneuvering targets in radar networking scenarios is studied in this *** the interacting multiple model algorithm and the expected-mode augmentation algorithm,the fixed base model set leads to a mismatch between the model set and the target motion mode,which causes the reduction on tracking *** adaptive grid-expected-mode augmentation variable structure multiple model algorithm is *** adaptive grid algorithm based on the turning model is extended to the two-dimensional pattern space to realize the self-adaptation of the model ***,combining with the unscented information filtering,and by interacting the measurement information of neighboring radars and iterating information matrix with consistency strategy,a distributed target tracking algorithm based on the posterior information of the information matrix is *** the problem of filtering divergence while target is leaving radar surveillance area,a k-coverage algorithm based on particle swarm optimization is applied to plan the radar motion trajectory for achieving filtering convergence.
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.
Visual localization is a crucial component in the application of mobile robot and autonomous *** retrieval is an efficient and effective technique in image-based localization *** to the drastic variability of environm...
详细信息
Visual localization is a crucial component in the application of mobile robot and autonomous *** retrieval is an efficient and effective technique in image-based localization *** to the drastic variability of environmental conditions,e.g.,illumination changes,retrievalbased visual localization is severely affected and becomes a challenging *** this work,a general architecture is first formulated probabilistically to extract domain-invariant features through multi-domain image ***,a novel gradientweighted similarity activation mapping loss(Grad-SAM)is incorporated for finer localization with high *** also propose a new adaptive triplet loss to boost the contrastive learning of the embedding in a self-supervised *** final coarse-to-fine image retrieval pipeline is implemented as the sequential combination of models with and without Grad-SAM *** experiments have been conducted to validate the effectiveness of the proposed approach on the CMU-Seasons *** strong generalization ability of our approach is verified with the RobotCar dataset using models pre-trained on urban parts of the CMU-Seasons *** performance is on par with or even outperforms the state-of-the-art image-based localization baselines in medium or high precision,especially under challenging environments with illumination variance,vegetation,and night-time ***,real-site experiments have been conducted to validate the efficiency and effectiveness of the coarse-to-fine strategy for localization.
This paper examines the event-triggered consensus of the multi-agent system on matrix-weighted networks, where the interdependencies among higher-dimensional states of neighboring agents are characterized by matrix-we...
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.
Learning-based multi-view stereo (MVS) method heavily relies on feature matching, which requires distinctive and descriptive representations. An effective solution is to apply non-local feature aggregation, e.g., Tran...
Learning-based multi-view stereo (MVS) method heavily relies on feature matching, which requires distinctive and descriptive representations. An effective solution is to apply non-local feature aggregation, e.g., Transformer. Albeit useful, these techniques introduce heavy computation overheads for MVS. Each pixel densely attends to the whole image. In contrast, we propose to constrain nonlocal feature augmentation within a pair of lines: each point only attends the corresponding pair of epipolar lines. Our idea takes inspiration from the classic epipolar geometry, which shows that one point with different depth hypotheses will be projected to the epipolar line on the other view. This constraint reduces the 2D search space into the epipolar line in stereo matching. Similarly, this suggests that the matching of MVS is to distinguish a series of points lying on the same line. Inspired by this point-toline search, we devise a line-to-point non-local augmentation strategy. We first devise an optimized searching algorithm to split the 2D feature maps into epipolar line pairs. Then, an Epipolar Transformer (ET) performs non-local feature augmentation among epipolar line pairs. We incorporate the ET into a learning-based MVS baseline, named ET-MVSNet. ET-MVSNet achieves state-of-the-art reconstruction performance on both the DTU and Tanks-and-Temples benchmark with high efficiency. Code is available at https://***/TQTQliu/ET-MVSNet.
Dear Editor,This letter proposes a process-monitoring method based on temporal feature agglomeration and enhancement,in which a novel feature extractor called contrastive feature extractor(CFE)extracts the temporal an...
详细信息
Dear Editor,This letter proposes a process-monitoring method based on temporal feature agglomeration and enhancement,in which a novel feature extractor called contrastive feature extractor(CFE)extracts the temporal and relational features among process *** the feature representations are enhanced by maximizing the separation among different classes while minimizing the scatter within each class.
暂无评论