相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Engineering, Electrical & Electronic
Wenjun Xu et al.
Summary: This article proposes a current decoupling control strategy by combining FTSMC and AESO. The voltage errors and external disturbances caused by d-q-axis current coupling and motor parameter variations are regarded as a lumped disturbance, facilitating the decoupling of d-q-axis currents. The FTSMC strategy is used to track the d-q-axis currents and improve the current dynamic performance, while the AESO method observes and compensates for the lumped disturbance.
IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS
(2023)
Article
Engineering, Electrical & Electronic
Haifeng Wang et al.
Summary: This article proposes a low-complexity model-predictive control method for multiphase open-end winding drive systems, which eliminates harmonic voltage vectors and reduces computational burden and complexity by synthesizing virtual vectors and compensating for dead time.
IEEE TRANSACTIONS ON POWER ELECTRONICS
(2022)
Article
Biology
Navdeep Kaur et al.
Summary: This paper presents a technique based on coattention and reinforcement learning for generating clinically accurate reports from chest x-ray images. The results demonstrate that the proposed model performs well for the given task and generates sufficiently accurate CXR reports.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Automation & Control Systems
Zhongbao Wei et al.
Summary: This article proposes a knowledge-based and multiphysics-constrained fast charging strategy for lithium-ion batteries, which takes into account thermal safety and degradation. The proposed strategy combines a model-based state observer with a deep reinforcement learning-based optimizer to provide a solution for fast charging. Experimental results demonstrate the superiority of the proposed strategy in terms of charging speed, thermal safety, and battery life extension.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Computer Science, Artificial Intelligence
Dong Xie et al.
Summary: In this article, a novel semicentralized deep deterministic policy gradient (SCDDPG) algorithm is proposed for cooperative multiagent games. The algorithm utilizes a two-level actor-critic structure to facilitate interactions and cooperation among agents in StarCraft combat. The local and global actor-critic structures work together to generate optimal control actions and improve cooperation in the games.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Automation & Control Systems
Dongyu Li et al.
Summary: The study introduces a novel control problem called time-synchronized stability (TSS) with unique finite/fixed-time stability considerations, presenting sufficient conditions for achieving (fixed-) TSS. It is found that norm-normalized sign functions contribute to achieving TSS in control system design. Additionally, a fixed-time-synchronized sliding-mode controller for second-order systems is proposed and singularity avoidance is considered.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2022)
Article
Chemistry, Analytical
Shiyun Liang et al.
Summary: This paper investigates a disturbance observer-based deep reinforcement learning control strategy to achieve high robustness and precise tracking performance in high-precision electromechanical systems. The proposed algorithm, embedded with an integral differential compensator and deep deterministic policy gradient, along with an adaptive sliding mode disturbance observer, demonstrates accurate target path tracking of the micropositioner.
Article
Multidisciplinary Sciences
Jun Wang et al.
Summary: In this study, a parallel robot equipped with a constant force actuator for grinding applications was designed. The characteristics of the robot's spatial positions and poses were analyzed using DH parameters and geometric methods. The workspace for grinding objects was found to be a cylindrical shape with a cross section similar to a symmetric circular sector. The forces produced by the robot system were analyzed using the Newton-Euler method, and the rationality of the force decoupling design was verified. The theoretical analyses were confirmed through theoretical calculations, simulations, and experimental analyses, providing a theoretical foundation for the design, manufacture, and control of the proposed parallel robot system.
Article
Computer Science, Artificial Intelligence
Dongmin Lee et al.
Summary: This paper explores the potential of deep reinforcement learning (DRL) for adaptive task allocation in dynamic robotic construction environments using a digital twin-driven learning method. The results of testing on a virtual robotic construction project showed that the DRL model's task allocation approach reduced construction time by 36% compared to a rule-based imperative model.
ADVANCED ENGINEERING INFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Qingxin Meng et al.
Summary: This article discusses an uncertain two-link rigid-flexible manipulator with vibration amplitude constraint, and aims to achieve its position control through motion planning and adaptive tracking approach. The motion trajectories planning for the manipulator's two links can guarantee reaching desired angles and suppress vibration, while the adaptive tracking controller enables the two links to track the planned trajectories under various uncertainties. Simulation results confirm the effectiveness of the proposed control strategy and the superior performance of motion planning and tracking controller.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Chemistry, Multidisciplinary
Jun Wang et al.
Summary: This paper proposes an algorithm for painting large objects based on a nine-axis UR5 robotic manipulator, aiming to improve the quality and efficiency of paint jobs. The algorithm consists of three phases: target point acquisition, trajectory planning, and UR5 robot inverse solution acquisition. The algorithm utilizes STL files, PCA algorithm, and k-d tree to obtain the point cloud model in the target point acquisition phase. Simulation results demonstrate the feasibility and effectiveness of the proposed algorithm.
APPLIED SCIENCES-BASEL
(2022)
Article
Biology
Mohammad Ali Raheb et al.
Summary: This study applied a reinforcement learning algorithm called normalized advantage function (NAF) to regulate blood glucose levels in type-2 diabetic patients. The results demonstrate the promising potential of this method in controlling blood glucose fluctuations.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Automation & Control Systems
Ruidong Yan et al.
Summary: A hybrid car-following strategy based on DDPG and CACC is proposed in this study to address the performance issues of DDPG in car-following, while maintaining the basic performance of CACC and utilizing the exploration advantages of DDPG on complex environments. Simulation results demonstrate an improvement in the car-following performance compared to DDPG and CACC.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2022)
Article
Energy & Fuels
Bing Sun et al.
Summary: This study proposes an optimization planning method considering wind and PV electricity curtailment measure, and solves the optimization model through the steady-state security region of the distribution network. Under the principle of economic consumption, the penetration rate of renewable energy electricity can be effectively improved.
Article
Mathematical & Computational Biology
Shuguang Li et al.
Summary: This paper considers the problem of minimizing makespan on uniform machines with restricted assignment. By eliminating the assumption, approximation algorithms with approximation ratios of 2 and 4/3 are presented for both inclusive and tree-hierarchical assignment restrictions.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2022)
Article
Mathematical & Computational Biology
Shuguang Li
Summary: This article discusses the problem of scheduling jobs with equal lengths on uniform parallel batch machines with non-identical capacities. It presents efficient exact algorithms for different objective functions and makespan minimization in cases of equal and unequal release times.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2022)
Article
Computer Science, Interdisciplinary Applications
Haeun Yoo et al.
Summary: Batch process control presents a challenge due to its dynamic operation over a large range of parameters. Nonlinear model predictive control (NMPC) is currently the standard for optimal control of batch processes but may be unsatisfactory in the presence of uncertainties. Reinforcement learning (RL) offers a viable alternative, but careful decisions on reward function design and value update method are necessary for effective application. This study introduces a phase segmentation approach for reward function design and value/policy function representation, along with modifications to the deep deterministic policy gradient algorithm (DDPG) to ensure more stable and efficient learning behavior. The proposed approach is demonstrated and further issues are highlighted through a case study on a batch polymerization process producing polyols.
COMPUTERS & CHEMICAL ENGINEERING
(2021)
Article
Computer Science, Interdisciplinary Applications
Xinjin Li et al.
Summary: The proposed hierarchical evacuation method utilizes a two-level evacuation mechanism to guide crowd evacuation, improving efficiency through path planning and collision avoidance. The method also leverages deep reinforcement learning algorithms to enhance learning efficiency.
COMPUTERS & INDUSTRIAL ENGINEERING
(2021)
Article
Biology
Hae-Ryong Yun et al.
Summary: A model for ESA dose recommendation was proposed for optimizing anemia management in patients with KFRT, based on sequential awareness neural networks. The model showed potential effectiveness in a simulated prospective study, achieving reductions in ESA dose, stable monthly Hb difference, and improved target Hb success rate compared to real-world clinical data.
COMPUTERS IN BIOLOGY AND MEDICINE
(2021)
Article
Engineering, Aerospace
Da Jiang et al.
Summary: This paper proposes a multiagent reinforcement learning approach to generate real-time inverse kinematic solutions for coordinated manipulators in response to collision avoidance and input saturation control issues. A competitive mechanism is developed through reasonable reward functions to maintain a safe distance between the dual arms. Simulation results show that this approach has higher accuracy in inverse kinematic trajectory planning.
JOURNAL OF AEROSPACE ENGINEERING
(2021)
Article
Energy & Fuels
Jiawen Li et al.
Summary: A novel automatic generation control dispatch was proposed to balance stochastic power disturbance in integrated energy system, aiming to reduce control deviation and regulation mileage payment while improving training efficiency and action quality through multiple experience pool probability replay strategy. The proposed algorithm was verified on a two-area load frequency control model and Hainan province IES for different energy demand.
Article
Computer Science, Artificial Intelligence
Shuai Han et al.
Summary: This paper introduces a new reinforcement learning algorithm RUD to address the inefficiency and instability of DDPG, demonstrating that RUD can better utilize new data and is more suitable for a specific strategy in terms of Q value variance. The experiments validate the effectiveness and superiority of RUD.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Automation & Control Systems
Zhizheng Zhang et al.
Summary: Asynchronous episodic DDPG (AE-DDPG) is an expansion of DDPG that achieves more effective learning through redesigned experience replay and a new type of noise in action space. Experiments demonstrate that AE-DDPG performs well in computationally complex environments and achieves higher rewards and sample efficiency in MuJoCo environments compared to other DDPG variants.
IEEE TRANSACTIONS ON CYBERNETICS
(2021)
Article
Chunyang HU et al.
Xibei Gongye Daxue Xuebao/Journal of Northwestern Polytechnical University
(2021)
Article
Computer Science, Interdisciplinary Applications
Tanuja Joshi et al.
Summary: Control of batch processes is a challenging task due to their complex dynamics and non-steady state operating conditions. Developing control strategies that directly interact with the process and learning from experiences can help address some of these challenges. The study introduces a novel actor-critic RL algorithm and demonstrates its efficacy in various batch process examples.
COMPUTERS & CHEMICAL ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Ruyi Dong et al.
Summary: The Kernel Search Optimization (KSO) algorithm was proposed to simplify the optimization process by transforming the optimization of nonlinear functions into a linear process. By adopting a local search of the hill-climbing algorithm and simplifying the calculation of kernel parameters, the improved algorithm outperformed the original KSO and some well-known algorithms in terms of accuracy and running time.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Chemistry, Multidisciplinary
Hiba Sekkat et al.
Summary: This study introduces a deep deterministic policy gradient approach for autonomous object grasping using a multi-degrees-of-freedom robotic arm through inverse kinematics. The approach demonstrates better performance compared to traditional methods and showcases its advantages through simulation results.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Information Systems
Luo Yang et al.
Summary: Learning a ball stroke strategy is crucial for accurate motion control in table tennis robots, and this study develops a stroke approach based on deep reinforcement learning with spin velocity estimation capability to achieve precise ball returns. By pre-training in a virtual table tennis environment and collecting simulated data, the proposed control strategy demonstrates superior performance compared to traditional methods in real robot implementation.
Article
Computer Science, Artificial Intelligence
Xiongtao Shi et al.
Article
Computer Science, Artificial Intelligence
Xiang Zhang et al.
EXPERT SYSTEMS WITH APPLICATIONS
(2020)
Article
Chemistry, Multidisciplinary
Ithan Moreira et al.
APPLIED SCIENCES-BASEL
(2020)
Article
Computer Science, Artificial Intelligence
Qian Shi et al.
Review
Computer Science, Information Systems
Hao-nan Wang et al.
FRONTIERS OF INFORMATION TECHNOLOGY & ELECTRONIC ENGINEERING
(2020)
Article
Engineering, Electrical & Electronic
Jingda Wu et al.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2020)
Article
Engineering, Electrical & Electronic
Yanchang Liang et al.
IEEE TRANSACTIONS ON POWER SYSTEMS
(2020)
Article
Computer Science, Information Systems
Chengrun Qiu et al.
IEEE INTERNET OF THINGS JOURNAL
(2019)
Article
Chemistry, Multidisciplinary
Ander Iriondo et al.
APPLIED SCIENCES-BASEL
(2019)
Article
Computer Science, Information Systems
Hongnan Liang et al.
Article
Ergonomics
Hatice Esen et al.
INTERNATIONAL JOURNAL OF OCCUPATIONAL SAFETY AND ERGONOMICS
(2019)
Article
Computer Science, Information Systems
Heming Jia et al.
Article
Engineering, Biomedical
En Lu et al.
JOURNAL OF VIBROENGINEERING
(2016)
Article
Engineering, Mechanical
Mohammad Reza Soltanpour et al.
NONLINEAR DYNAMICS
(2013)
Review
Automation & Control Systems
Juan M. Martin-Sanchez et al.
INTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING
(2012)
