检索结果-内蒙古大学图书馆

Learning control barrier functions from expert demonstrations

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Robey, Alexander Hu, Haimin Lindemann, Lars Zhang, Hanwen Dimarogonas, Dimos V. Tu, Stephen Matni, Nikolai Department of Electrical and Systems Engineering University of Pennsylvania Division of Decision and Control Systems KTH Royal Institute of Technology Google Brain Robotics

Inspired by the success of imitation and inverse reinforcement learning in replicating expert behavior through optimal control, we propose a learning based approach to safe controller synthesis based on control barrier functions (CBFs). We consider the setting of a known nonlinear control affine dynamical system and assume that we have access to safe trajectories generated by an expert a practical example of such a setting would be a kinematic model of a self-driving vehicle with safe trajectories (e.g., trajectories that avoid collisions with obstacles in the environment) generated by a human driver. We then propose and analyze an optimization-based approach to learning a CBF that enjoys provable safety guarantees under suitable Lipschitz smoothness assumptions on the underlying dynamical system. A strength of our approach is that it is agnostic to the parameterization used to represent the CBF, assuming only that the Lipschitz constant of such functions can be efficiently bounded. Furthermore, if the CBF parameterization is convex, then under mild assumptions, so is our learning process. We end with extensive numerical evaluations of our results on both planar and realistic examples, using both random feature and deep neural network parameterizations of the CBF. To the best of our knowledge, these are the first results that learn provably safe control barrier functions from data. Copyright © 2020, The Authors. All rights reserved.

关键词： Parameterization

Temperature-Dependent Modeling of Thermoelectric Elements

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Evers, Enzo Slenders, Rens van Gils, Rob Oomen, Tom Control Systems Technology Mechanical Engineering Department Eindhoven University of Technology Eindhoven Netherlands Philips Innovation Services Eindhoven Netherlands

Active thermal control is crucial in achieving the required accuracy and throughput in many industrial applications, e.g., in the medical industry, high-power lighting industry, and semiconductor industry. Thermoelectric Modules (TEMs) can be used to both heat and cool, alleviating some of the challenges associated with traditional electric heater based control. However, the dynamic behavior of these modules is non-affine in their inputs and state, complicating their implementation. To facilitate advanced control approaches a high fidelity model is required. In this work an approach is presented that increases the modeling accuracy by incorporating temperature dependent parameters. Using an experimental identification procedure, the parameters are estimated under different operating conditions. The resulting model achieves superior accuracy for a wide range of temperatures, demonstrated using experimental validation measurements. Copyright © 2020, The Authors. All rights reserved.

关键词： Semiconductor device manufacture

Distributed Optimal Allocation with Quantized Communication and Privacy-Preserving Guarantees

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Nylöf, Jakob Rikos, Apostolos I. Gracy, Sebin Johansson, Karl Henrik Division of Decision and Control Systems School of Electrical Engineering and Computer Science KTH Royal Institute of Technology and Digital Futures Stockholm Sweden Department of Electrical and Computer Engineering Rice University HoustonTX United States

In this paper, we analyze the problem of optimally allocating resources in a distributed and privacy-preserving manner. We propose a novel distributed optimal resource allocation algorithm with privacy-preserving guarantees, which operates over a directed communication network. Our algorithm converges in finite time and allows each node to process and transmit quantized messages. Our algorithm utilizes a distributed quantized average consensus strategy combined with a privacy-preserving mechanism. We show that the algorithm converges in finite-time, and we prove that, under specific conditions on the network topology, nodes are able to preserve the privacy of their initial state. Finally, to illustrate the results, we consider an example where test kits need to be optimally allocated proportionally to the number of infections in a region. It is shown that the proposed privacy-preserving resource allocation algorithm performs well with an appropriate convergence rate under privacy guarantees. © 2021, CC BY.

关键词： Resource allocation

Fusing online gaussian process-based learning and control for scanning quantum dot microscopy

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Pfefferkorn, Maik Maiworm, Michael Wagner, Christian Tautz, F. Stefan Findeisen, Rolf Otto-von-Guericke-Universitaet Magdeburg Laboratory for Systems Theory and Automatic Control Germany Juelich Research Center Juelich Germany Fundamentals of Future Information Technology Juelich Germany

Elucidating electrostatic surface potentials contributes to a deeper understanding of the nature of matter and its physicochemical properties, which is the basis for a wide field of applications. Scanning quantum dot microscopy, a recently developed technique allows to measure such potentials with atomic resolution. For an efficient deployment in scientific practice, however, it is essential to speed up the scanning process. To this end we employ a two-degree-of-freedom control paradigm, in which a Gaussian process is used as the feedforward part. We present a tailored online learning scheme of the Gaussian process, adapted to scanning quantum dot microscopy, that includes hyperparameter optimization during operation to enable fast and precise scanning of arbitrary surface structures. For the potential application in practice, the accompanying computational cost is reduced evaluating different sparse approximation approaches. The fully independent training conditional approximation, used on a reduced set of active training data, is found to be the most promising approach. Copyright © 2020, The Authors. All rights reserved.

关键词： Scanning

Evaluation of Path States of Large IoT Networks Using Locally and Remotely controlled Measurements

学校读者我要写书评

暂无评论

Evaluation of Path States of Large IoT Networks Using Locall...

IFIP Networking Conference

作者： Agnieszka Chodorek Robert R. Chodorek Faculty of Electrical Engineering Automatic Control and Computer Science Kielce University of Technology Kielce Poland Department of Telecommunications The AGH University of Science and Technology Kraków Poland

ISBN: (数字)9783903176287

ISBN: (纸本)9781728167107

The state of network paths is of great importance for the quality and reliability of the transmission of information coming from the IoT devices. Changing network conditions cause the state of a single path to be highly unpredictable, especially when large-scale IoT networks, connecting thousands of specialized devices, are in use. The solution to this problem is to perform trials in a real environment, with the use of real hardware. The rationalization of the design, the implementation and the exploitation of the process will be the main benefit of this approach. In this demonstration, a testbed infrastructure for the estimation of the state of path parameters in wired and wireless networks is presented. The estimation is based mainly on active measurements performed with the use of the Probing Packet Trains method. The key elements of the testbed infrastructure are locally and remotely controlled energy-efficient single-board computers, working under the control of the Linux operating system, which are equipped with software able to track (in manual or automatic mode) changing conditions of a given transmission path.

关键词： Estimation Internet of Things Monitoring Performance evaluation Servers Manuals Bandwidth

Deep attributed network representation learning via attribute enhanced neighborhood

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Li, Cong Shi, Min Qu, Bo Li, Xiang The Adaptive Networks and Control Lab Department of Electronic Engineering The Research Center of Smart Networks and Systems School of Information Science and Technology Fudan University Shanghai200433 China Peng Cheng Laboratory Shenzhen518000 China

Attributed network representation learning aims at learning node embeddings by integrating network structure and attribute information. It is a challenge to fully capture the microscopic structure and the attribute semantics simultaneously, where the microscopic structure includes the one-step, two-step and multi-step relations, indicating the first-order, second-order and high-order proximity of nodes, respectively. In this paper, we propose a deep attributed network representation learning via attribute enhanced neighborhood (DANRL-ANE) model to improve the robustness and effectiveness of node representations. The DANRL-ANE model adopts the idea of the autoencoder, and expands the decoder component to three branches to capture different order proximity. We linearly combine the adjacency matrix with the attribute similarity matrix as the input of our model, where the attribute similarity matrix is calculated by the cosine similarity between the attributes based on the social homophily. In this way, we preserve the second-order proximity to enhance the robustness of DANRL-ANE model on sparse networks, and deal with the topological and attribute information simultaneously. Moreover, the sigmoid cross-entropy loss function is extended to capture the neighborhood character, so that the first-order proximity is better preserved. We compare our model with the state-of-the-art models on five real-world datasets and two network analysis tasks, i.e., link prediction and node classification. The DANRL-ANE model performs well on various networks, even on sparse networks or networks with isolated nodes given the attribute information is sufficient. Copyright © 2021, The Authors. All rights reserved.

关键词： Classification (of information)

Multi timescale battery modeling: Integrating physics insights to data-driven model

学校读者我要写书评

暂无评论

Applied Energy 2025年 393卷

作者： Desai, Tushar Gallo, Alexander J. Ferrari, Riccardo M.G. Delft Center for Systems and Control Delft University of Technology Mekelweg 2 South Holland DelftCD2628 Netherlands Department of Electronics Information and Bioengineering Politecnico di Milano Via Ponzio 34 Milano20133 Italy

Developing accurate models for batteries, capturing ageing effects and nonlinear behaviors, is critical for the development of efficient and effective performance. Due to the inherent difficulties in developing physics-based models, data-driven techniques have been gaining popularity. However, most machine learning methods are black boxes, lacking interpretability and requiring large amounts of labeled data. In this paper, we propose a physics-informed encoder–decoder model that learns from unlabeled data to separate slow-changing battery states, such as state of charge (SOC) and state of health (SOH), from fast transient responses, thereby increasing interpretability compared to conventional methods. By integrating physics-informed loss functions and modified architectures, we map the encoder output to quantifiable battery states, without needing explicit SOC and SOH labels. Our proposed approach is validated on a lithium-ion battery ageing dataset capturing dynamic discharge profiles that aim to mimic electric vehicle driving profiles. The model is trained and validated on sparse intermittent cycles (6 %–7 % of all cycles), accurately estimating SOC and SOH while providing accurate multistep ahead voltage predictions across single and multiple-cell based training scenarios. © 2025 The Author(s)

关键词： State of charge

From Bipedal Walking to Quadrupedal Locomotion: Full-Body Dynamics Decomposition for Rapid Gait Generation

学校读者我要写书评

暂无评论

From Bipedal Walking to Quadrupedal Locomotion: Full-Body Dy...

IEEE International Conference on Robotics and Automation (ICRA)

作者： Wen-Loong Ma Aaron D. Ames department of mechanical engineering California Institute of Technology Pasadena CA faculty of the department of Control and Dynamical Systems California Institute of Technology Pasadena CA

ISBN: (数字)9781728173955

ISBN: (纸本)9781728173962

This paper systematically decomposes a quadrupedal robot into bipeds to rapidly generate walking gaits and then recomposes these gaits to obtain quadrupedal locomotion. We begin by decomposing the full-order, nonlinear and hybrid dynamics of a three-dimensional quadrupedal robot, including its continuous and discrete dynamics, into two bipedal systems that are subject to external forces. Using the hybrid zero dynamics (HZD) framework, gaits for these bipedal robots can be rapidly generated (on the order of seconds) along with corresponding controllers. The decomposition is achieved in such a way that the bipedal walking gaits and controllers can be composed to yield dynamic walking gaits for the original quadrupedal robot - the result is the rapid generation of dynamic quadruped gaits utilizing the full-order dynamics. This methodology is demonstrated through the rapid generation (3.96 seconds on average) of four stepping-in-place gaits and one diagonally symmetric ambling gait at 0.35 m/s on a quadrupedal robot - the Vision 60, with 36 state variables and 12 control inputs - both in simulation and through outdoor experiments. This suggested a new approach for fast quadrupedal trajectory planning using full-body dynamics, without the need for empirical model simplification, wherein methods from dynamic bipedal walking can be directly applied to quadrupeds.

关键词： Legged locomotion Robot kinematics Dynamics Nonlinear dynamical systems Jacobian matrices Trajectory

Performance Analysis of Multi-RIS-Aided LoRa systems With Outdated and Imperfect CSI

学校读者我要写书评

暂无评论

IEEE Transactions on Communications 2024年

作者： Liang, Zhaokun Cai, Guofa He, Jiguang Kaddoum, Georges Huang, Chongwen Guangdong University of Technology School of Information Engineering China Ministry of Education Key Laboratory of Industrial Internet of Things and Networked Control Chongqing University of Posts and Telecommunications Chongqing400065 China Great Bay University School of Computing and Information Technology China University of Oulu Centre for Wireless Communications Oulu90014 Finland University of Quebec Electrical Engineering Department École de Technologie Supérieure MontrealQCH2L 2C4 Canada Lebanese American University Artificial Intelligence and Cyber Systems Research Center United States Zhejiang University College of Information Science and Electronic Engineering Hangzhou310027 China

Although LoRa has emerged as the leading technology among the rapidly developing low-power wide-area networks, the performance of the LoRa system severely deteriorates over fading channels. To address this problem, in this paper, we introduce multiple reconfigurable intelligent surfaces (multi-RISs) into the LoRa system to improve its performance. Our specific focus is on the impact of outdated channel state information (CSI), the imperfection of estimated CSI, and the design of RIS discrete phase shifts on the performance. To this end, we first use the moment-matching method to obtain the end-to-end (E2E) channel coefficient of the joint outdated channels and erroneous channels over Nakagami-m fading. Moreover, the closed-form bit error rates (BERs) of the proposed system with non-coherent and coherent detections are derived. The results reveal that, in the high signal-to-noise ratio (SNR) regime, coherent detection encounters the error floor and performs worse than non-coherent detection. Furthermore, we also analyze delay outage rate, throughput, and achievable diversity order of the proposed system. The results show that, despite the presence of outdated CSI and channel estimation errors, the proposed system is still superior to RIS-aided LoRa systems adopting blind transmission and RIS-free ones. Finally, we also thoroughly investigate the effects of various important factors such as the correlation factor, channel estimation errors, the number of RIS reflecting elements, and the number of quantization bits for RIS discrete phase shifts on the performance. © 1972-2012 IEEE.

关键词： Fading channels