检索结果-内蒙古大学图书馆

A Fault Isolation Approach for Switched Systems Based on ADT Method

学校读者我要写书评

暂无评论

A Fault Isolation Approach for Switched Systems Based on ADT...

IEEE Conference on Industrial Electronics and Applications (ICIEA)

作者： Shafqat Ali Yuchen Jiang Hao Luo Faizan Shahid Muhammad Taskeen Raza Shah Faisal Control and Simulation Center Harbin Institute of Technology Harbin China Department of Electrical Engineering Lahore College for Women University Lahore Pakistan

ISBN: (数字)9798350360868

ISBN: (纸本)9798350360875

Ensuring safety and reliability in switched systems depends on precise control and maintaining stability across different operational modes, thereby guaranteeing fault-free performance and minimizing the risk of system failures. This paper presents a technique for detecting and isolating sensor faults in switched linear systems with an average dwell time switching approach. To achieve this, a series of sliding mode observers are designed to generate a bank of residual signals, which helps to isolate the sensor faults based on their corresponding residuals. The proposed scheme ensures stability analysis of the switched system by considering the Lyapunov function and the average dwell time switching constraints, and the $H$ ∞ performance index achieves robustness from faults to the residual error signal. Sufficient conditions are established using linear matrix inequalities and feasible solutions for the designed scheme. Finally, a buck-boost converter simulation is presented to showcase the efficiency of the proposed method.

关键词： Switched systems Linear systems Sufficient conditions Stability criteria Switches Observers Safety

Enhanced Preprocessing and Fast PCA Feature Extraction Algorithm for Ultrasound Images of Ovarian Diseases

学校读者我要写书评

暂无评论

Enhanced Preprocessing and Fast PCA Feature Extraction Algor...

International Conference on Signal Processing Proceedings (ICSP)

作者： Yue Zhao Hongbei Xiang Kuo Miao Xiaoqiu Dong Zhiyu Zhou Dong Zhu Control Theory and Engineering School of Astronautics Harbin Institute of Technology Harbin China Department of Ultrasound Fourth Affiliated Hospital of Harbin Medical University Harbin China Chengdu CHG Robots & Intelligent EquipmentInstitute Co. Ltd Chengdu China

ISBN: (数字)9798350387384

ISBN: (纸本)9798350387391

Clinical diagnosis of ovarian diseases is mainly based on ultrasonography. Due to the susceptibility of ultrasound images to noise, and the special characteristics and defects of ovarian ultrasound images themselves are prominent, there are some limitations in the existing preprocessing and feature extraction methods. To address the above problems, this paper proposes an enhanced preprocessing and fast principal component analysis (PCA) feature extraction algorithm for ovarian ultrasound images. Combining the unique characteristics of ovarian ultrasound images, the image preprocessing algorithm is designed using a strategy that combines image sharpening using Robert’s operator with linear transformation of contrast. The traditional PCA algorithm is improved to reduce and extract ovarian image features more effectively. The experimental results demonstrate that the proposed method can achieve a generalization rate of 89.79%, AUC of 0.9136 and a duration of 1.98s.

关键词： Measurement Ultrasonic imaging Noise Signal processing algorithms Ultrasonography Feature extraction Image preprocessing Lesions Principal component analysis Diseases

Exploring Adversarial Robustness of Deep State Space Models

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Qi, Biqing Luo, Yiang Gao, Junqi Li, Pengfei Tian, Kai Ma, Zhiyuan Zhou, Bowen Department of Electronic Engineering Tsinghua University China Shanghai Artificial Intelligence Laboratory China Department of Control Science and Engineering Harbin Institute of Technology China School of Mathematics Harbin Institute of Technology China

Deep State Space Models (SSMs) have proven effective in numerous task scenarios but face significant security challenges due to Adversarial Perturbations (APs) in real-world deployments. Adversarial Training (AT) is a mainstream approach to enhancing Adversarial Robustness (AR) and has been validated on various traditional DNN architectures. However, its effectiveness in improving the AR of SSMs remains unclear. While many enhancements in SSM components, such as integrating Attention mechanisms and expanding to data-dependent SSM parameterizations, have brought significant gains in Standard Training (ST) settings, their potential benefits in AT remain unexplored. To investigate this, we evaluate existing structural variants of SSMs with AT to assess their AR performance. We observe that pure SSM structures struggle to benefit from AT, whereas incorporating Attention yields a markedly better trade-off between robustness and generalization for SSMs in AT compared to other components. Nonetheless, the integration of Attention also leads to Robust Overfitting (RO) issues. To understand these phenomena, we empirically and theoretically analyze the output error of SSMs under AP. We find that fixed-parameterized SSMs have output error bounds strictly related to their parameters, limiting their AT benefits, while input-dependent SSMs may face the problem of error explosion. Furthermore, we show that the Attention component effectively scales the output error of SSMs during training, enabling them to benefit more from AT, but at the cost of introducing RO due to its high model complexity. Inspired by this, we propose a simple and effective Adaptive Scaling (AdS) mechanism that brings AT performance close to Attention-integrated SSMs without introducing the issue of RO. Our code is available at Robustness-of-SSM. © 2024, CC BY-SA.

关键词： State space methods

Coverage control for Multi-Agent Systems via Curve Arc Length

学校读者我要写书评

暂无评论

Coverage Control for Multi-Agent Systems via Curve Arc Lengt...

Chinese Automation Congress (CAC)

作者： Kaiqi Yang Xiaoli Wang Hengyu Yang Wei Xie Department of Control Science and Engineering Harbin Institute of Technology Harbin China School of Information Science and Engineering Harbin Institute of Technology at Weihai Weihai China School of Aeronautics and Astronautics Tiangong University Tianjin China

ISBN: (数字)9798350368604

ISBN: (纸本)9798350368611

In complex environments, effective and comprehensive information gathering necessitates the strategic deployment of multi-agent systems. Corridor-type regions, characterized as non-convex areas, present substantial challenges in formulating coverage control strategies. To tackle this issue, this paper proposes a novel coverage control methodology for multi-agent systems based on partitioning the coverage area using curve arc lengths. Firstly, the coverage areas for each team is segmented into sequentially arranged subregions based on curve arc lengths. Through adjustments in the virtual leading agent's position, the range of curve arc lengths for each team's coverage areas is dynamically managed. Then, a distributed coverage controller is developed for agents, which integrates the movement status of virtual leading agents to guide agents towards the centroid positions of their respective coverage areas. Finally, numerical simulations affirm the efficacy of the proposed coverage control approach.

关键词： Automation control systems Numerical simulation Multi-agent systems

Less is More: Efficient Model Merging with Binary Task Switch

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Qi, Biqing Li, Fangyuan Wang, Zhen Gao, Junqi Li, Dong Ye, Peng Zhou, Bowen Shanghai Artificial Intelligence Laboratory China Department of Control Science and Engineering Harbin Institute of Technology China School of Mathematics Harbin Institute of Technology China Department of Electronic Engineering Tsinghua University China

As an effective approach to equip models with multi-task capabilities without additional training, model merging has garnered significant attention. However, existing methods face challenges of redundant parameter conflicts and the excessive storage burden of parameters. In this work, through controlled experiments, we reveal that for task vectors, only those parameters with magnitudes above a certain threshold contribute positively to the task, exhibiting a pulse-like characteristic. We then attempt leveraging this characteristic to binarize the task vectors and reduce storage overhead. Further controlled experiments show that the binarized task vectors incur almost no decrease in fine-tuning and merging performance, and even exhibit stronger performance improvements as the proportion of redundant parameters increases. Based on these insights, we propose Task Switch (T-Switch), which decomposes task vectors into three components: 1) an activation switch instantiated by a binarized mask vector, 2) a polarity switch instantiated by a binarized sign vector, and 3) a scaling knob instantiated by a scalar coefficient. By storing task vectors in a binarized form, T-Switch alleviates parameter conflicts while ensuring efficient task parameter storage. Furthermore, to enable automated switch combination in T-Switch, we further introduce Auto-Switch, which enables training-free switch combination via retrieval from a small query set. Experiments indicate that our methods achieve significant performance improvements over existing baselines, requiring only 1-3% of the storage space of full-precision parameters. © 2024, CC BY.

关键词： Vectors

Evolution of Thought: Diverse and High-Quality Reasoning via Multi-Objective Optimization

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Qi, Biqing Qian, Zhouyi Luo, Yiang Gao, Junqi Li, Dong Zhang, Kaiyan Zhou, Bowen Shanghai Artificial Intelligence Laboratory China School of Mathematics Harbin Institute of Technology China Department of Control Science and Engineering Harbin Institute of Technology China Department of Electronic Engineering Tsinghua University China

As multi-modal large language models (MLLMs) are increasingly applied to complex reasoning tasks, the diversity and quality of reasoning paths become crucial factors affecting their performance. Although current methods aim to enhance reasoning quality through path expansion, they often neglect the diversity of reasoning paths and effective information sharing, leading to local optima and inefficiency. To address these challenges, we propose Evolution of Thought (EoT), a multi-objective framework designed to improve reasoning by fostering both high-quality and diverse reasoning paths. Specifically, we introduce the Non-dominated Sorting Genetic Algorithm II for multi-objective optimization, utilizing crossover and mutation operators to promote greater diversity in reasoning solutions. Additionally, we propose a Condensation-Aggregation mechanism to cluster and eliminate redundant paths, facilitate improved information sharing among parent nodes, and ultimately enhance both the efficiency and quality of the reasoning process. Validation experiments on various vision-language and language reasoning tasks demonstrate that EoT achieves superior reasoning performance and efficiency compared to other competitive baselines. Our study provides a novel perspective on the design of heuristic reasoning frameworks for MLLMs. © 2024, CC BY-SA.

关键词： Visual languages

SMR: State Memory Replay for Long Sequence Modeling

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Qi, Biqing Gao, Junqi Zhang, Kaiyan Li, Dong Liu, Jianxing Wu, Ligang Zhou, Bowen Department of Control Science and Engineering Harbin Institute of Technology China Department of Electronic Engineering Tsinghua University China School of Mathematics Harbin Institute of Technology China Frontis.AI Beijing China

Despite the promising performance of state space models (SSMs) in long sequence modeling, limitations still exist. Advanced SSMs like S5 and S6 (Mamba) in addressing non-uniform sampling, their recursive structures impede efficient SSM computation via convolution. To overcome compatibility limitations in parallel convolutional computation, this paper proposes a novel non-recursive non-uniform sample processing strategy. Theoretical analysis of SSMs through the lens of Event-Triggered control (ETC) theory reveals the Non-Stable State (NSS) problem, where deviations from sampling point requirements lead to error transmission and accumulation, causing the divergence of the SSM’s hidden state. Our analysis further reveals that adjustments of input sequences with early memories can mitigate the NSS problem, achieving Sampling Step Adaptation (SSA). Building on this insight, we introduce a simple yet effective plug-and-play mechanism, State Memory Replay (SMR), which utilizes learnable memories to adjust the current state with multi-step information for generalization at sampling points different from those in the training data. This enables SSMs to stably model varying sampling points. Experiments on long-range modeling tasks in autoregressive language modeling and Long Range Arena demonstrate the general effectiveness of the SMR mechanism for a series of SSM models. © 2024, CC BY.

关键词： State space methods

Perturbation Event Oriented Dynamic Task Scheduling Algorithm for Parallel Testing of control Systems

学校读者我要写书评

暂无评论

Perturbation Event Oriented Dynamic Task Scheduling Algorith...

International Conference on Sensing, Diagnostics, Prognostics, and control (SDPC)

作者： Jingyi Chen Shenhang Wang Xiaoxuan Tong Gang Xiang Yang Yu Ziyuan Zhan School of Electronic and Information Engineeer Harbin Institute of Technology Harbin China System Design Department Beijing Aerospace Automatic Control Institute Beijing China Beijing Aerospace Automatic Control Institute Beijing China Department of System Engineering Beijing Aerospace Automatic Control Institute Beijing China

ISBN: (数字)9798350388855

ISBN: (纸本)9798350388862

Task scheduling algorithms are key to research and practice in parallel testing techniques, especially in control system applications. The complexity of control system testing requires higher scheduling efficiency. However, perturbation events caused by test equipment problems, data transmission obstacles, or task changes are almost unavoidable and may reduce the scheduling effectiveness or even lead to system crashes. Therefore, in this paper, based on the analysis of the dynamic test scheduling model, a disturbance factor is designed to judge the urgency of disturbance events for evaluating the dynamic scheduling cycle. On this basis, based on the comprehensive rescheduling index, a suitable adjustment strategy is selected to formulate the dynamic test scheduling scheme. The experimental results show that the algorithm proposed in this paper can achieve fast response capability for different disturbance events and exhibit good adaptivity.

关键词： Scheduling algorithms Perturbation methods Heuristic algorithms Dynamic scheduling control systems Computer crashes Sensors Test equipment Indexes Data communication

A time-delay approach to stabilization by extremum seeking controller via simplified ISS analysis

学校读者我要写书评

暂无评论

IFAC-PapersOnLine 2023年第2期56卷 4289-4294页

作者： Jin Zhang Emilia Fridman Xuefei Yang School of Mechatronic Engineering and Automation Shanghai University Shanghai 200072 China School of Electrical Engineering Tel Aviv University Tel Aviv 69978 Israel Center for Control Theory and Guidance Technology Harbin Institute of Technology Harbin China

We study stabilization of linear uncertain systems under unknown control directions using a bounded extremum seeking controller with a small parameter. We consider a small time-varying measurement delay. By using the recently proposed time-delay approach to Lie-Brackets-based averaging, we transform the closed-loop system to a time-delay (neutral type) one, which has a form of perturbed Lie brackets system. The input-to-state stability (ISS) of the time-delay system guarantees the same for the original one. Differently from the existing analysis via Lyapunov-Krasovskii (L-K) method, we transform the neutral system to an ordinary differential equation (ODE) with delayed perturbations and employ variation of constants formula that greatly simplifies the analysis and leads to simpler stability conditions. Two numerical examples illustrate that the proposed method allows essentially larger parameter uncertainties with bounds on the small parameter and time-delay that are not too small.

关键词： Time-delay Lie-brackets-based averaging extremum seeking