检索结果-内蒙古大学图书馆

Multi-Point Positional Insertion Tuning for Small Object Detection

学校读者我要写书评

暂无评论

Multi-Point Positional Insertion Tuning for Small Object Det...

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Kanoko Goto Takumi Karasawa Takumi Hirose Rei Kawakami Nakamasa Inoue Dept. of Computer Science Institute of Science Tokyo Tokyo Japan Dept. of Systems and Control Engineering Institute of Science Tokyo Tokyo Japan

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Small object detection aims to localize and classify small objects within images. With recent advances in large-scale vision-language pretraining, finetuning pretrained object detection models has emerged as a promising approach. However, finetuning large models is computationally and memory expensive. To address this issue, this paper introduces multi-point positional insertion (MPI) tuning, a parameter-efficient finetuning (PEFT) method for small object detection. Specifically, MPI incorporates multiple positional embeddings into a frozen pretrained model, enabling the efficient detection of small objects by providing precise positional information to latent features. Through experiments, we demonstrated the effectiveness of the proposed method on the SODA-D dataset. MPI performed comparably to conventional PEFT methods, including CoOp and VPT, while significantly reducing the number of parameters that need to be tuned.

关键词： Computational modeling Object detection Signal processing Feature extraction Encoding Acoustics Speech processing Tuning

Bayes and Biased Estimators Without Hyper-parameter Estimation: Comparable Performance to the Empirical-Bayes-Based Regularized Estimator

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Ju, Yue Wahlberg, Bo Hjalmarsson, Håkan Division of Decision and Control Systems School of Electrical Engineering and Computer Science KTH Royal Institute of Technology Stockholm10044 Sweden

Regularized system identification has become a significant complement to more classical system identification. It has been numerically shown that kernel-based regularized estimators often perform better than the maximum likelihood estimator in terms of minimizing mean squared error (MSE). However, regularized estimators often require hyper-parameter estimation. This paper focuses on ridge regression and the regularized estimator by employing the empirical Bayes hyperparameter estimator. We utilize the excess MSE to quantify the MSE difference between the empirical-Bayes-based regularized estimator and the maximum likelihood estimator for large sample sizes. We then exploit the excess MSE expressions to develop both a family of generalized Bayes estimators and a family of closed-form biased estimators. They have the same excess MSE as the empirical-Bayes-based regularized estimator but eliminate the need for hyper-parameter estimation. Moreover, we conduct numerical simulations to show that the performance of these new estimators is comparable to the empirical-Bayes-based regularized estimator, while computationally, they are more efficient. Copyright © 2025, The Authors. All rights reserved.

关键词： Mean square error

On Space-Filling Input Design for Nonlinear Dynamic Model Learning: A Gaussian Process Approach

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Liu, Yuhan Kiss, Máté Tóth, Roland Schoukens, Maarten Control Systems Group Department of Electrical Engineering Eindhoven University of Technology Eindhoven5600 MB Netherlands Systems and Control Laboratory HUN-REN Institute for Computer Science and Control Budapest1111 Hungary

While optimal input design for linear systems has been well-established, no systematic approach exists for nonlinear systems, where robustness to extrapolation/interpolation errors is prioritized over minimizing estimated parameter variance. To address this issue, we develop a novel identification-oriented space-filling input design strategy for efficient covering of a well-defined region of interest. By placing a Gaussian Process (GP) prior on the joint input-state space, the proposed strategy leverages the GP posterior variance to construct a space-filling promoting input design cost function from the Bayesian point of view. This accordingly enables maximization of the coverage in the region of interest, thereby facilitating the generation of informative datasets. Furthermore, we theoretically prove that minimization of the cost function indicates the space-filling behavior. The effectiveness of the proposed strategy is demonstrated on both an academic example and a mass-spring-damper system, highlighting its potential practical impact on efficient exploration in identification scenarios for nonlinear dynamical systems. © 2025, CC BY-NC-SA.

关键词： Cost functions

Equilibrium-Independent control of Continuous-Time Nonlinear Systems Via the LPV Framework

学校读者我要写书评

暂无评论

IEEE Transactions on Automatic control 2025年

作者： Koelewijn, Patrick J. W. Weiland, Siep Toth, Roland Eindhoven University of Technology Control Systems Group Department of Electrical Engineering Eindhoven Netherlands Institute for Computer Science and Control Systems and Control Laboratory Budapest Hungary

In this paper, we consider the analysis and control of continuous-time nonlinear systems to ensure universal shifted stability and performance, i.e., stability and performance w.r.t. each forced equilibrium point of the system. This "equilibrium-free"concept is especially beneficial for control problems that require the tracking of setpoints and rejection of persistent disturbances, such as input loads. In this paper, we show how the velocity form, i.e., the time-differentiated dynamics of the system, plays a crucial role in characterizing these properties and how the analysis of it can be solved by the application of Linear Parameter-Varying (LPV) methods in a computationally efficient manner. Furthermore, by leveraging the properties of the velocity form and the LPV framework, a novel controller synthesis method is presented which ensures closed-loop universal shifted stability and performance. The proposed controller design is verified experimentally on a real system. Additionally, we compare the proposed method to a standard LPV control design, demonstrating the improved stability and performance guarantees of the new approach. © 1963-2012 IEEE.

关键词： control system synthesis

A third-party platoon coordination service: Pricing under government subsidies

学校读者我要写书评

暂无评论

A third-party platoon coordination service: Pricing under go...

作者： Bai, Ting Johansson, Alexander Li, Shaoyuan Johansson, Karl Henrik Mårtensson, Jonas School of Electrical Engineering and Computer Science KTH Royal Institute of Technology Stockholm Sweden KTH Royal Institute of Technology Stockholm Sweden Department of Automation Shanghai Jiao Tong University Shanghai China Key Laboratory of System Control and Information Processing Shanghai Jiao Tong University Shanghai China Digital Futures Stockholm Sweden

This paper models a platooning system consisting of trucks and a third-party service provider (TPSP), which performs platoon coordination, distributes the platooning profit in platoons, and charges trucks in exchange for the services. Government subsidies used to incentivize platooning are also considered. We propose a pricing rule for the TPSP, which keeps part of the platooning profit including the subsidy each time a platoon is formed. In addition, a platoon coordination solution based on the distributed model predictive control (MPC) is proposed, in which the pricing rule under government subsidies is integrated. We perform a realistic simulation over the Swedish road network to evaluate the impact of the pricing rule and subsidies on the achieved profits and fuel savings. Our results show that subsidies are an effective mean to boost fuel savings from platooning. Moreover, the simulation study indicates that high pricing corresponds to a low platooning rate of the system, as trucks' incentives for platooning decrease. © 2023 The Authors. Asian Journal of control published by John Wiley & Sons Australia, Ltd on behalf of Chinese Automatic control Society.

关键词： Trucks

Distributed model predictive control without terminal cost under inexact distributed optimization

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Liu, Xiaoyu Dimarogonas, Dimos V. Liu, Changxin Dabiri, Azita De Schutter, Bart Delft Center for Systems and Control Delft University of Technology Netherlands Division of Decision and Control Systems School of Electrical Engineering and Computer Science KTH Royal Institute of Technology Sweden

This paper presents a novel distributed model predictive control (MPC) formulation without terminal cost and a corresponding distributed synthesis approach for distributed linear discrete-time systems with coupled constraints. The proposed control scheme introduces an explicit stability condition as an additional constraint based on relaxed dynamic programming. As a result, contrary to other related approaches, system stability with the developed controller does not rely on designing a terminal cost. A distributed synthesis approach is then introduced to handle the stability constraint locally within each local agent. To solve the underlying optimization problem for distributed MPC, a violation-free distributed optimization approach is developed, using constraint tightening to ensure feasibility throughout iterations. A numerical example demonstrates that the proposed distributed MPC approach ensures closed-loop stability for each feasible control sequence, with each agent computing its control input in parallel. © 2025, CC BY.

关键词： Predictive control systems

Enriching Physical-Virtual Interaction in AR Gaming by Tracking Identical Real Objects

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Yu, Liuchuan Huang, Ching-I Wang, Hsueh-Cheng Yu, Lap-Fai Computer Science Department George Mason University FairfaxVA United States Institute of Electrical and Control Engineering National Yang Ming Chiao Tung University Taiwan Faculty of Electrical and Computer Engineering Department and Institute of Electrical and Control Engineering National Yang Ming Chiao Tung University Taiwan Faculty of Computer Science Department George Mason University FairfaxVA United States

Augmented reality (AR) games, particularly those designed for headsets, have become increasingly prevalent with advancements in both hardware and software. However, the majority of AR games still rely on pre-scanned or static scenes, and interaction mechanisms are often limited to controllers or hand-tracking. Additionally, the presence of identical objects in AR games poses challenges for conventional object tracking techniques, which often struggle to differentiate between identical objects or necessitate the installation of fixed cameras for global object movement tracking. In response to these limitations, we present a novel approach to address the tracking of identical objects in an AR scene to enrich physical-virtual interaction. Our method leverages partial scene observations captured by an AR headset, utilizing the perspective and spatial data provided by this technology. Object identities within the scene are determined through the solution of a label assignment problem using integer programming. To enhance computational efficiency, we incorporate a Voronoi diagram-based pruning method into our approach. Our implementation of this approach in a farm-to-table AR game demonstrates its satisfactory performance and robustness. Furthermore, we showcase the versatility and practicality of our method through applications in AR storytelling and a simulated gaming robot. Our video demo is available at: https://***/ rPGkLYuKvCQ. Copyright © 2025, The Authors. All rights reserved.

关键词： Computational efficiency

Binary Stochastic Flip Optimization for Training Binary Neural Networks

学校读者我要写书评

暂无评论

Binary Stochastic Flip Optimization for Training Binary Neur...

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Tatsukichi Shibuya Nakamasa Inoue Rei Kawakami Ikuro Sato Department of Computer Science Institute of Science Tokyo Tokyo Japan Department of Systems and Control Engineering Institute of Science Tokyo Tokyo Japan Denso IT Laboratory Tokyo Japan

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

For deploying deep neural networks on edge devices with limited resources, binary neural networks (BNNs) have attracted significant attention, due to their computational and memory efficiency. However, once a neural network is binarized, finetuning it on edge devices becomes challenging because most conventional training algorithms for BNNs are designed for use on centralized servers and require storing real-valued parameters during training. To address this limitation, this paper introduces binary stochastic flip optimization (BinSFO), a novel training algorithm for BNNs. BinSFO employs a parameter update rule based on Boolean operations, eliminating the need to store real-valued parameters and thereby reducing memory requirements and computational overhead. In experiments, we demonstrated the effectiveness and memory efficiency of BinSFO in fine-tuning scenarios on six image classification datasets. BinSFO performed comparably to conventional training algorithms with a 70.7% smaller memory requirement. Code is released at https://***/TatsukichiShibuya/ICASSP2025_BinSFO

关键词： Training Speech coding Memory management Signal processing algorithms Signal processing Classification algorithms Servers Speech processing Optimization Image classification

Towards backward compatibility of ADRC: revisiting classical state-feedback control with integral compensator

学校读者我要写书评

暂无评论

Bulletin of the Polish Academy of Sciences: Technical Sciences 2025年第2期73卷

作者： Mrotek, Mikolaj Michalski, Jacek Madonski, Rafal Pazderski, Dariusz Retinger, Marek Faculty of Control Robotics and Electrical Engineering Institute of Robotics and Machine Intelligence Poznan University of Technology Piotrowo 3a Poznan60-965 Poland Faculty of Automatic Control Electronics and Computer Science Department of Automatic Control and Robotics Silesian University of Technology Akademicka 16 Gliwice44-100 Poland Faculty of Control Robotics and Electrical Engineering Institute of Automatic Control and Robotics Poznan University of Technology Piotrowo 3a Poznan60-965 Poland

In this paper, the problem of backward compatibility of active disturbance rejection control (ADRC) is investigated. The goal is to contextualize ADRC to deliver its interpretations from the established field of linear control systems. For this study, a control algorithm, denoted here as integral disturbance rejection control (IDRC), is considered that combines classical state-feedback control with an integral compensator. At first, an interpretation of ADRC is involved in terms of existing state-space control approaches. Next, a transition to the frequency domain is performed, which is justified as a significant part of practical control engineering is conducted in that domain. For assumed specific plant structures, both ADRC and IDRC are then holistically compared in terms of transfer function representation and frequency characteristics, as well as steady-state convergence conditions. Such a juxtaposition helps to highlight the similarities and differences of both approaches, whereas the utilized bandwidth parameterization is shown to bring the control system to the same form, thus indicating some interesting practical aspects. Finally, the theoretical results concerning both considered control structures are validated in a set of numerical simulations and experiments conducted on a laboratory hardware testbed. © 2025 The Author(s).

关键词： Disturbance rejection