The wireless communication technologies have fundamentally revolutionized industrial operations. The operation of the automated equipment is conducted in a closed-loop manner, where the status of devices is collected and sent to the control center through the uplink channel, and the control center sends the calculated control commands back to the devices via downlink communication. However, existing studies neglect the interdependent relationship between uplink and downlink communications, and there is an absence of a unified approach to model the communication, sensing, and control within the loop. This can lead to inaccurate performance assessments, ultimately hindering the ability to provide guidance for the design of practical systems. Therefore, this paper introduces an integrated closed-loop model that encompasses sensing, communication, and control functionalities, while addressing the coupling effects between uplink and downlink communications. Through the analysis of system convergence, an inequality pertaining to the performances of sensing, communication, and control is derived. Additionally, a joint optimization algorithm for control and resource allocation is proposed. Simulation results are presented to offer an intuitive understanding of the impact of system parameters. The findings of this paper unveil the intricate correlation among sensing, communication, and control, providing insights for the optimal design of industrial closed-loop systems.
{"title":"Communication, Sensing and Control integrated Closed-loop System: Modeling, Control Design and Resource Allocation","authors":"Zeyang Meng, Dingyou Ma, Zhiqing Wei, Ying Zhou, Zhiyong Feng","doi":"arxiv-2409.11796","DOIUrl":"https://doi.org/arxiv-2409.11796","url":null,"abstract":"The wireless communication technologies have fundamentally revolutionized\u0000industrial operations. The operation of the automated equipment is conducted in\u0000a closed-loop manner, where the status of devices is collected and sent to the\u0000control center through the uplink channel, and the control center sends the\u0000calculated control commands back to the devices via downlink communication.\u0000However, existing studies neglect the interdependent relationship between\u0000uplink and downlink communications, and there is an absence of a unified\u0000approach to model the communication, sensing, and control within the loop. This\u0000can lead to inaccurate performance assessments, ultimately hindering the\u0000ability to provide guidance for the design of practical systems. Therefore,\u0000this paper introduces an integrated closed-loop model that encompasses sensing,\u0000communication, and control functionalities, while addressing the coupling\u0000effects between uplink and downlink communications. Through the analysis of\u0000system convergence, an inequality pertaining to the performances of sensing,\u0000communication, and control is derived. Additionally, a joint optimization\u0000algorithm for control and resource allocation is proposed. Simulation results\u0000are presented to offer an intuitive understanding of the impact of system\u0000parameters. The findings of this paper unveil the intricate correlation among\u0000sensing, communication, and control, providing insights for the optimal design\u0000of industrial closed-loop systems.","PeriodicalId":501175,"journal":{"name":"arXiv - EE - Systems and Control","volume":"112 2 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mehdi Heydari Shahna, Pauli Mustalahti, Jouni Mattila
In-wheel drive (IWD) systems enhance the responsiveness, traction, and maintenance efficiency of vehicles by enabling each wheel to operate independently. This paper proposes a novel robust torque-observed valve-based control (RTOVC) framework to address velocity tracking in hydraulic IWDs that actuate heavy-duty wheeled mobile robots (HWMRs), considering such challenges as wheel slippages, sensor limitations, rough terrains, and modeling uncertainties. To overcome the sensor-dependent control systems associated with the closed-loop torque/pressure in hydraulic IWD-actuated HWMRs, a robust observer network based on an adaptive barrier Lyapunov function (BLF) is proposed to estimate the required in-wheel motor torque to track the velocity references. Then, another adaptive BLF for valve control signals is employed to modulate the hydraulic fluid to generate the estimated torque for each IWD. The RTOVC strategy ensures user-defined safety within the logarithmic BLF framework by constraining the valve control signal, actual velocity, velocity tracking error, and torque of each hydraulic IWD in an HWMR to avoid exceeding specified limits. Despite its safety constraints, external disturbances, and modeling uncertainties, robustness and uniformly exponential stability of the RTOVC-applied hydraulic IWD mechanism are ensured in HWMRs. Experimental investigations using a 6,500-kg HWMR, actuated by four independent IWDs under intense disturbances and safety-defined constraints, validate the performance of the RTOVC.
{"title":"Robust Sensor-Limited Control with Safe Input-Output Constraints for Hydraulic In-Wheel Motor Drive Mobility Systems","authors":"Mehdi Heydari Shahna, Pauli Mustalahti, Jouni Mattila","doi":"arxiv-2409.11823","DOIUrl":"https://doi.org/arxiv-2409.11823","url":null,"abstract":"In-wheel drive (IWD) systems enhance the responsiveness, traction, and\u0000maintenance efficiency of vehicles by enabling each wheel to operate\u0000independently. This paper proposes a novel robust torque-observed valve-based\u0000control (RTOVC) framework to address velocity tracking in hydraulic IWDs that\u0000actuate heavy-duty wheeled mobile robots (HWMRs), considering such challenges\u0000as wheel slippages, sensor limitations, rough terrains, and modeling\u0000uncertainties. To overcome the sensor-dependent control systems associated with\u0000the closed-loop torque/pressure in hydraulic IWD-actuated HWMRs, a robust\u0000observer network based on an adaptive barrier Lyapunov function (BLF) is\u0000proposed to estimate the required in-wheel motor torque to track the velocity\u0000references. Then, another adaptive BLF for valve control signals is employed to\u0000modulate the hydraulic fluid to generate the estimated torque for each IWD. The\u0000RTOVC strategy ensures user-defined safety within the logarithmic BLF framework\u0000by constraining the valve control signal, actual velocity, velocity tracking\u0000error, and torque of each hydraulic IWD in an HWMR to avoid exceeding specified\u0000limits. Despite its safety constraints, external disturbances, and modeling\u0000uncertainties, robustness and uniformly exponential stability of the\u0000RTOVC-applied hydraulic IWD mechanism are ensured in HWMRs. Experimental\u0000investigations using a 6,500-kg HWMR, actuated by four independent IWDs under\u0000intense disturbances and safety-defined constraints, validate the performance\u0000of the RTOVC.","PeriodicalId":501175,"journal":{"name":"arXiv - EE - Systems and Control","volume":"40 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mohammad Bahari, Alvaro Paz, Mehdi Heydari Shahna, Jouni Mattila
The global push for sustainability and energy efficiency is driving significant advancements across various industries, including the development of electrified solutions for heavy-duty mobile manipulators (HDMMs). Electromechanical linear actuators (EMLAs), powered by permanent magnet synchronous motors, present an all-electric alternative to traditional internal combustion engine (ICE)-powered hydraulic actuators, offering a promising path toward an eco-friendly future for HDMMs. However, the limited operational range of electrified HDMMs, closely tied to battery capacity, highlights the need to fully exploit the potential of EMLAs that driving the manipulators. This goal is contingent upon a deep understanding of the harmonious interplay between EMLA mechanisms and the dynamic behavior of heavy-duty manipulators. To this end, this paper introduces a bilevel multi-objective optimization framework, conceptualizing the EMLA-actuated manipulator of an electrified HDMM as a leader--follower scenario. At the leader level, the optimization algorithm maximizes EMLA efficiency by considering electrical and mechanical constraints, while the follower level optimizes manipulator motion through a trajectory reference generator that adheres to manipulator limits. This optimization approach ensures that the system operates with a synergistic trade-off between the most efficient operating region of the actuation system, achieving a total efficiency of 70.3%, and high manipulator performance. Furthermore, to complement this framework and ensure precise tracking of the generated optimal trajectories, a robust, adaptive, subsystem-based control strategy is developed with accurate control and exponential stability. The proposed methodologies are validated on a three-degrees-of-freedom manipulator, demonstrating significant efficiency improvements while maintaining high-performance operation.
{"title":"System-Level Efficient Performance of EMLA-Driven Heavy-Duty Manipulators via Bilevel Optimization Framework with a Leader--Follower Scenario","authors":"Mohammad Bahari, Alvaro Paz, Mehdi Heydari Shahna, Jouni Mattila","doi":"arxiv-2409.11849","DOIUrl":"https://doi.org/arxiv-2409.11849","url":null,"abstract":"The global push for sustainability and energy efficiency is driving\u0000significant advancements across various industries, including the development\u0000of electrified solutions for heavy-duty mobile manipulators (HDMMs).\u0000Electromechanical linear actuators (EMLAs), powered by permanent magnet\u0000synchronous motors, present an all-electric alternative to traditional internal\u0000combustion engine (ICE)-powered hydraulic actuators, offering a promising path\u0000toward an eco-friendly future for HDMMs. However, the limited operational range\u0000of electrified HDMMs, closely tied to battery capacity, highlights the need to\u0000fully exploit the potential of EMLAs that driving the manipulators. This goal\u0000is contingent upon a deep understanding of the harmonious interplay between\u0000EMLA mechanisms and the dynamic behavior of heavy-duty manipulators. To this\u0000end, this paper introduces a bilevel multi-objective optimization framework,\u0000conceptualizing the EMLA-actuated manipulator of an electrified HDMM as a\u0000leader--follower scenario. At the leader level, the optimization algorithm\u0000maximizes EMLA efficiency by considering electrical and mechanical constraints,\u0000while the follower level optimizes manipulator motion through a trajectory\u0000reference generator that adheres to manipulator limits. This optimization\u0000approach ensures that the system operates with a synergistic trade-off between\u0000the most efficient operating region of the actuation system, achieving a total\u0000efficiency of 70.3%, and high manipulator performance. Furthermore, to\u0000complement this framework and ensure precise tracking of the generated optimal\u0000trajectories, a robust, adaptive, subsystem-based control strategy is developed\u0000with accurate control and exponential stability. The proposed methodologies are\u0000validated on a three-degrees-of-freedom manipulator, demonstrating significant\u0000efficiency improvements while maintaining high-performance operation.","PeriodicalId":501175,"journal":{"name":"arXiv - EE - Systems and Control","volume":"92 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yunfan Gao, Florian Messerer, Niels van Duijkeren, Boris Houska, Moritz Diehl
Online planning of collision-free trajectories is a fundamental task for robotics and self-driving car applications. This paper revisits collision avoidance between ellipsoidal objects using differentiable constraints. Two ellipsoids do not overlap if and only if the endpoint of the vector between the center points of the ellipsoids does not lie in the interior of the Minkowski sum of the ellipsoids. This condition is formulated using a parametric over-approximation of the Minkowski sum, which can be made tight in any given direction. The resulting collision avoidance constraint is included in an optimal control problem (OCP) and evaluated in comparison to the separating-hyperplane approach. Not only do we observe that the Minkowski-sum formulation is computationally more efficient in our experiments, but also that using pre-determined over-approximation parameters based on warm-start trajectories leads to a very limited increase in suboptimality. This gives rise to a novel real-time scheme for collision-free motion planning with model predictive control (MPC). Both the real-time feasibility and the effectiveness of the constraint formulation are demonstrated in challenging real-world experiments.
{"title":"Real-Time-Feasible Collision-Free Motion Planning For Ellipsoidal Objects","authors":"Yunfan Gao, Florian Messerer, Niels van Duijkeren, Boris Houska, Moritz Diehl","doi":"arxiv-2409.12007","DOIUrl":"https://doi.org/arxiv-2409.12007","url":null,"abstract":"Online planning of collision-free trajectories is a fundamental task for\u0000robotics and self-driving car applications. This paper revisits collision\u0000avoidance between ellipsoidal objects using differentiable constraints. Two\u0000ellipsoids do not overlap if and only if the endpoint of the vector between the\u0000center points of the ellipsoids does not lie in the interior of the Minkowski\u0000sum of the ellipsoids. This condition is formulated using a parametric\u0000over-approximation of the Minkowski sum, which can be made tight in any given\u0000direction. The resulting collision avoidance constraint is included in an\u0000optimal control problem (OCP) and evaluated in comparison to the\u0000separating-hyperplane approach. Not only do we observe that the Minkowski-sum\u0000formulation is computationally more efficient in our experiments, but also that\u0000using pre-determined over-approximation parameters based on warm-start\u0000trajectories leads to a very limited increase in suboptimality. This gives rise\u0000to a novel real-time scheme for collision-free motion planning with model\u0000predictive control (MPC). Both the real-time feasibility and the effectiveness\u0000of the constraint formulation are demonstrated in challenging real-world\u0000experiments.","PeriodicalId":501175,"journal":{"name":"arXiv - EE - Systems and Control","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Differential Dynamic Programming (DDP) is one of the indirect methods for solving an optimal control problem. Several extensions to DDP have been proposed to add stagewise state and control constraints, which can mainly be classified as augmented lagrangian methods, active set methods, and barrier methods. In this paper, we use an interior point method, which is a type of barrier method, to incorporate arbitrary stagewise equality and inequality state and control constraints. We also provide explicit update formulas for all the involved variables. Finally, we apply this algorithm to example systems such as the inverted pendulum, a continuously stirred tank reactor, car parking, and obstacle avoidance.
{"title":"Differential dynamic programming with stagewise equality and inequality constraints using interior point method","authors":"Siddharth Prabhu, Srinivas Rangarajan, Mayuresh Kothare","doi":"arxiv-2409.12048","DOIUrl":"https://doi.org/arxiv-2409.12048","url":null,"abstract":"Differential Dynamic Programming (DDP) is one of the indirect methods for\u0000solving an optimal control problem. Several extensions to DDP have been\u0000proposed to add stagewise state and control constraints, which can mainly be\u0000classified as augmented lagrangian methods, active set methods, and barrier\u0000methods. In this paper, we use an interior point method, which is a type of\u0000barrier method, to incorporate arbitrary stagewise equality and inequality\u0000state and control constraints. We also provide explicit update formulas for all\u0000the involved variables. Finally, we apply this algorithm to example systems\u0000such as the inverted pendulum, a continuously stirred tank reactor, car\u0000parking, and obstacle avoidance.","PeriodicalId":501175,"journal":{"name":"arXiv - EE - Systems and Control","volume":"31 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Bernard T. AgyemanUniversity of Alberta, Jinfeng LiuUniversity of Alberta, Sirish L. Shah
Efficient water management in agriculture is important for mitigating the growing freshwater scarcity crisis. Mixed-integer Model Predictive Control (MPC) has emerged as an effective approach for addressing the complex scheduling problems in agricultural irrigation. However, the computational complexity of mixed-integer MPC still poses a significant challenge, particularly in large-scale applications. This study proposes an approach to enhance the computational efficiency of mixed-integer MPC-based irrigation schedulers by employing ReLU surrogate models to describe the soil moisture dynamics of the agricultural field. By leveraging the mixed-integer linear representation of the ReLU operator, the proposed approach transforms the mixed-integer MPC-based scheduler with a quadratic cost function into a mixed-integer quadratic program, which is the simplest class of mixed-integer nonlinear programming problems that can be efficiently solved using global optimization solvers. The effectiveness of this approach is demonstrated through comparative studies conducted on a large-scale agricultural field across two growing seasons, involving other machine learning surrogate models, specifically Long Short-Term Memory (LSTM) networks, and the widely used triggered irrigation scheduling method. The ReLU-based approach significantly reduces solution times -- by up to 99.5% -- while achieving comparable performance to the LSTM approach in terms of water savings and Irrigation Water Use Efficiency (IWUE). Moreover, the ReLU-based approach maintains enhanced performance in terms of total prescribed irrigation and IWUE compared to the widely-used triggered irrigation scheduling method.
{"title":"ReLU Surrogates in Mixed-Integer MPC for Irrigation Scheduling","authors":"Bernard T. AgyemanUniversity of Alberta, Jinfeng LiuUniversity of Alberta, Sirish L. Shah","doi":"arxiv-2409.12082","DOIUrl":"https://doi.org/arxiv-2409.12082","url":null,"abstract":"Efficient water management in agriculture is important for mitigating the\u0000growing freshwater scarcity crisis. Mixed-integer Model Predictive Control\u0000(MPC) has emerged as an effective approach for addressing the complex\u0000scheduling problems in agricultural irrigation. However, the computational\u0000complexity of mixed-integer MPC still poses a significant challenge,\u0000particularly in large-scale applications. This study proposes an approach to\u0000enhance the computational efficiency of mixed-integer MPC-based irrigation\u0000schedulers by employing ReLU surrogate models to describe the soil moisture\u0000dynamics of the agricultural field. By leveraging the mixed-integer linear\u0000representation of the ReLU operator, the proposed approach transforms the\u0000mixed-integer MPC-based scheduler with a quadratic cost function into a\u0000mixed-integer quadratic program, which is the simplest class of mixed-integer\u0000nonlinear programming problems that can be efficiently solved using global\u0000optimization solvers. The effectiveness of this approach is demonstrated\u0000through comparative studies conducted on a large-scale agricultural field\u0000across two growing seasons, involving other machine learning surrogate models,\u0000specifically Long Short-Term Memory (LSTM) networks, and the widely used\u0000triggered irrigation scheduling method. The ReLU-based approach significantly\u0000reduces solution times -- by up to 99.5% -- while achieving comparable\u0000performance to the LSTM approach in terms of water savings and Irrigation Water\u0000Use Efficiency (IWUE). Moreover, the ReLU-based approach maintains enhanced\u0000performance in terms of total prescribed irrigation and IWUE compared to the\u0000widely-used triggered irrigation scheduling method.","PeriodicalId":501175,"journal":{"name":"arXiv - EE - Systems and Control","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper explores the potential application of Deep Reinforcement Learning in the furniture industry. To offer a broad product portfolio, most furniture manufacturers are organized as a job shop, which ultimately results in the Job Shop Scheduling Problem (JSSP). The JSSP is addressed with a focus on extending traditional models to better represent the complexities of real-world production environments. Existing approaches frequently fail to consider critical factors such as machine setup times or varying batch sizes. A concept for a model is proposed that provides a higher level of information detail to enhance scheduling accuracy and efficiency. The concept introduces the integration of DRL for production planning, particularly suited to batch production industries such as the furniture industry. The model extends traditional approaches to JSSPs by including job volumes, buffer management, transportation times, and machine setup times. This enables more precise forecasting and analysis of production flows and processes, accommodating the variability and complexity inherent in real-world manufacturing processes. The RL agent learns to optimize scheduling decisions. It operates within a discrete action space, making decisions based on detailed observations. A reward function guides the agent's decision-making process, thereby promoting efficient scheduling and meeting production deadlines. Two integration strategies for implementing the RL agent are discussed: episodic planning, which is suitable for low-automation environments, and continuous planning, which is ideal for highly automated plants. While episodic planning can be employed as a standalone solution, the continuous planning approach necessitates the integration of the agent with ERP and Manufacturing Execution Systems. This integration enables real-time adjustments to production schedules based on dynamic changes.
{"title":"Optimizing Job Shop Scheduling in the Furniture Industry: A Reinforcement Learning Approach Considering Machine Setup, Batch Variability, and Intralogistics","authors":"Malte Schneevogt, Karsten Binninger, Noah Klarmann","doi":"arxiv-2409.11820","DOIUrl":"https://doi.org/arxiv-2409.11820","url":null,"abstract":"This paper explores the potential application of Deep Reinforcement Learning\u0000in the furniture industry. To offer a broad product portfolio, most furniture\u0000manufacturers are organized as a job shop, which ultimately results in the Job\u0000Shop Scheduling Problem (JSSP). The JSSP is addressed with a focus on extending\u0000traditional models to better represent the complexities of real-world\u0000production environments. Existing approaches frequently fail to consider\u0000critical factors such as machine setup times or varying batch sizes. A concept\u0000for a model is proposed that provides a higher level of information detail to\u0000enhance scheduling accuracy and efficiency. The concept introduces the\u0000integration of DRL for production planning, particularly suited to batch\u0000production industries such as the furniture industry. The model extends\u0000traditional approaches to JSSPs by including job volumes, buffer management,\u0000transportation times, and machine setup times. This enables more precise\u0000forecasting and analysis of production flows and processes, accommodating the\u0000variability and complexity inherent in real-world manufacturing processes. The\u0000RL agent learns to optimize scheduling decisions. It operates within a discrete\u0000action space, making decisions based on detailed observations. A reward\u0000function guides the agent's decision-making process, thereby promoting\u0000efficient scheduling and meeting production deadlines. Two integration\u0000strategies for implementing the RL agent are discussed: episodic planning,\u0000which is suitable for low-automation environments, and continuous planning,\u0000which is ideal for highly automated plants. While episodic planning can be\u0000employed as a standalone solution, the continuous planning approach\u0000necessitates the integration of the agent with ERP and Manufacturing Execution\u0000Systems. This integration enables real-time adjustments to production schedules\u0000based on dynamic changes.","PeriodicalId":501175,"journal":{"name":"arXiv - EE - Systems and Control","volume":"12 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264196","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Microgrids (MGs) have been equipped with large-scale distributed energy sources (DESs), and become more vulnerable due to the low inertia characteristic. In particular, high-density misbehaving DESs caused by cascading faults bring a great challenge to frequency synchronization and active power sharing among DESs. To tackle the problem, we propose a fully distributed resilient consensus protocol, which utilizes confidence weights to evaluate the level of trust among agents with a first-order filter and a softmax-type function. We pioneer the analysis of this nonlinear control system from the system operating range and the graph structure perspectives. Both necessary and sufficient conditions are provided to ensure DACC to be uniformly ultimately bounded, even in a robust network with low connectivity. Simulations on a modified IEEE33-bus microgrid testbed with 17 DESs validate that DACC outperforms existing methods in the presence of 8 misbehaving DESs.
微电网（MGs）配备了大规模分布式能源（DESs），由于其低惰性的特点而变得更加脆弱。特别是由级联故障引起的高密度误动作分布式电源给频率同步和分布式电源之间的有功功率共享带来了巨大挑战。为了解决这个问题，我们提出了一种全分布式弹性共识协议，它利用置信度权重，通过一阶滤波器和软极大值函数来评估代理之间的信任程度。我们率先从系统运行范围和图结构的角度分析了这一非线性控制系统。我们提供了必要条件和充分条件，以确保 DACC 即使在连通性较低的鲁棒网络中也能均匀终界。在具有 17 个 DES 的改进型 IEEE33 总线微电网试验平台上进行的仿真验证了 DACC 在存在 8 个行为不端的 DES 的情况下性能优于现有方法。
{"title":"Distributed Resilient Secondary Control for Microgrids with Attention-based Weights against High-density Misbehaving Agents","authors":"Yutong Li, Lili Wang","doi":"arxiv-2409.11812","DOIUrl":"https://doi.org/arxiv-2409.11812","url":null,"abstract":"Microgrids (MGs) have been equipped with large-scale distributed energy\u0000sources (DESs), and become more vulnerable due to the low inertia\u0000characteristic. In particular, high-density misbehaving DESs caused by\u0000cascading faults bring a great challenge to frequency synchronization and\u0000active power sharing among DESs. To tackle the problem, we propose a fully\u0000distributed resilient consensus protocol, which utilizes confidence weights to\u0000evaluate the level of trust among agents with a first-order filter and a\u0000softmax-type function. We pioneer the analysis of this nonlinear control system\u0000from the system operating range and the graph structure perspectives. Both\u0000necessary and sufficient conditions are provided to ensure DACC to be uniformly\u0000ultimately bounded, even in a robust network with low connectivity. Simulations\u0000on a modified IEEE33-bus microgrid testbed with 17 DESs validate that DACC\u0000outperforms existing methods in the presence of 8 misbehaving DESs.","PeriodicalId":501175,"journal":{"name":"arXiv - EE - Systems and Control","volume":"9 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Juan Cervino, Saurav Agarwal, Vijay Kumar, Alejandro Ribeiro
The multi-objective coverage control problem requires a robot swarm to collaboratively provide sensor coverage to multiple heterogeneous importance density fields (IDFs) simultaneously. We pose this as an optimization problem with constraints and study two different formulations: (1) Fair coverage, where we minimize the maximum coverage cost for any field, promoting equitable resource distribution among all fields; and (2) Constrained coverage, where each field must be covered below a certain cost threshold, ensuring that critical areas receive adequate coverage according to predefined importance levels. We study the decentralized setting where robots have limited communication and local sensing capabilities, making the system more realistic, scalable, and robust. Given the complexity, we propose a novel decentralized constrained learning approach that combines primal-dual optimization with a Learnable Perception-Action-Communication (LPAC) neural network architecture. We show that the Lagrangian of the dual problem can be reformulated as a linear combination of the IDFs, enabling the LPAC policy to serve as a primal solver. We empirically demonstrate that the proposed method (i) significantly outperforms existing state-of-the-art decentralized controllers by 30% on average in terms of coverage cost, (ii) transfers well to larger environments with more robots and (iii) is scalable in the number of fields and robots in the swarm.
{"title":"Constrained Learning for Decentralized Multi-Objective Coverage Control","authors":"Juan Cervino, Saurav Agarwal, Vijay Kumar, Alejandro Ribeiro","doi":"arxiv-2409.11311","DOIUrl":"https://doi.org/arxiv-2409.11311","url":null,"abstract":"The multi-objective coverage control problem requires a robot swarm to\u0000collaboratively provide sensor coverage to multiple heterogeneous importance\u0000density fields (IDFs) simultaneously. We pose this as an optimization problem\u0000with constraints and study two different formulations: (1) Fair coverage, where\u0000we minimize the maximum coverage cost for any field, promoting equitable\u0000resource distribution among all fields; and (2) Constrained coverage, where\u0000each field must be covered below a certain cost threshold, ensuring that\u0000critical areas receive adequate coverage according to predefined importance\u0000levels. We study the decentralized setting where robots have limited\u0000communication and local sensing capabilities, making the system more realistic,\u0000scalable, and robust. Given the complexity, we propose a novel decentralized\u0000constrained learning approach that combines primal-dual optimization with a\u0000Learnable Perception-Action-Communication (LPAC) neural network architecture.\u0000We show that the Lagrangian of the dual problem can be reformulated as a linear\u0000combination of the IDFs, enabling the LPAC policy to serve as a primal solver.\u0000We empirically demonstrate that the proposed method (i) significantly\u0000outperforms existing state-of-the-art decentralized controllers by 30% on\u0000average in terms of coverage cost, (ii) transfers well to larger environments\u0000with more robots and (iii) is scalable in the number of fields and robots in\u0000the swarm.","PeriodicalId":501175,"journal":{"name":"arXiv - EE - Systems and Control","volume":"32 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Active Inference is a framework that emphasizes the interaction between agents and their environment. While the framework has seen significant advancements in the development of agents, the environmental models are often borrowed from reinforcement learning problems, which may not fully capture the complexity of multi-agent interactions or allow complex, conditional communication. This paper introduces Reactive Environments, a comprehensive paradigm that facilitates complex multi-agent communication. In this paradigm, both agents and environments are defined as entities encapsulated by boundaries with interfaces. This setup facilitates a robust framework for communication in nonequilibrium-Steady-State systems, allowing for complex interactions and information exchange. We present a Julia package RxEnvironments.jl, which is a specific implementation of Reactive Environments, where we utilize a Reactive Programming style for efficient implementation. The flexibility of this paradigm is demonstrated through its application to several complex, multi-agent environments. These case studies highlight the potential of Reactive Environments in modeling sophisticated systems of interacting agents.
主动推理是一种强调代理与其环境之间互动的框架。虽然该框架在代理开发方面取得了重大进展，但环境模型通常借鉴自强化学习问题，可能无法完全捕捉到多代理交互的复杂性，也不允许复杂的条件通信。本文介绍了 "反应式环境"（Reactive Environments），这是一种促进复杂多代理交流的综合范式。在这一范式中，代理和环境都被定义为由带有接口的边界封装的实体。这种设置为非平衡-稳态系统中的通信提供了一个稳健的框架，允许复杂的交互和信息交换。我们提出了一个 Julia 包 RxEnvironments.jl，它是反应式环境的具体实现，我们利用反应式编程风格来高效地实现它。通过将其应用于多个复杂的多代理环境，我们展示了这一范式的灵活性。这些案例研究凸显了反应式环境在模拟复杂的交互代理系统方面的潜力。
{"title":"Reactive Environments for Active Inference Agents with RxEnvironments.jl","authors":"Wouter W. L. Nuijten, Bert de Vries","doi":"arxiv-2409.11087","DOIUrl":"https://doi.org/arxiv-2409.11087","url":null,"abstract":"Active Inference is a framework that emphasizes the interaction between\u0000agents and their environment. While the framework has seen significant\u0000advancements in the development of agents, the environmental models are often\u0000borrowed from reinforcement learning problems, which may not fully capture the\u0000complexity of multi-agent interactions or allow complex, conditional\u0000communication. This paper introduces Reactive Environments, a comprehensive\u0000paradigm that facilitates complex multi-agent communication. In this paradigm,\u0000both agents and environments are defined as entities encapsulated by boundaries\u0000with interfaces. This setup facilitates a robust framework for communication in\u0000nonequilibrium-Steady-State systems, allowing for complex interactions and\u0000information exchange. We present a Julia package RxEnvironments.jl, which is a\u0000specific implementation of Reactive Environments, where we utilize a Reactive\u0000Programming style for efficient implementation. The flexibility of this\u0000paradigm is demonstrated through its application to several complex,\u0000multi-agent environments. These case studies highlight the potential of\u0000Reactive Environments in modeling sophisticated systems of interacting agents.","PeriodicalId":501175,"journal":{"name":"arXiv - EE - Systems and Control","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：