2024
|
Sun, Haining; Tang, Xiaoqiang; Ge, Shuzhi Sam Controllable Trade-Off Between Performance and Constrained Input for Vibration Suppression in Flexible Space Structures IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS, 60 (3), pp. 3390-3402, 2024, DOI: 10.1109/TAES.2024.3361434. Abstract | BibTeX | Endnote @article{ISI:001246582400007,
title = {Controllable Trade-Off Between Performance and Constrained Input for Vibration Suppression in Flexible Space Structures},
author = {Haining Sun and Xiaoqiang Tang and Shuzhi Sam Ge},
doi = {10.1109/TAES.2024.3361434},
times_cited = {0},
issn = {0018-9251},
year = {2024},
date = {2024-06-01},
journal = {IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS},
volume = {60},
number = {3},
pages = {3390-3402},
publisher = {IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC},
address = {445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA},
abstract = {Achieving vibration suppression of flexible space structures requires not only considering the vibration settling time, but also avoiding excessive control forces that might damage the structure. In this article, an active control scheme is designed to suppress undesired vibrations in flexible structures by exerting small cable forces. It employs a cable-driven parallel robot (CDPR) as an actuator. The prominent feature of the controller lies in its ability to achieve a controllable trade-off between the effectiveness of vibration suppression and control inputs. Moreover, the controller can effectively suppress vibrations even when cable forces are constrained. The underlying principle is that the vibration energy is consumed through the negative work done by the cable forces until the tip displacement is ultimately reduced to a small range. The stability of the controller is verified by the Lyapunov method. Simulations demonstrate the effectiveness of the proposed control scheme, while experimental validation on a prototype consisting of a six-meter-long flexible structure and a four-cable CDPR further supports these findings.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Achieving vibration suppression of flexible space structures requires not only considering the vibration settling time, but also avoiding excessive control forces that might damage the structure. In this article, an active control scheme is designed to suppress undesired vibrations in flexible structures by exerting small cable forces. It employs a cable-driven parallel robot (CDPR) as an actuator. The prominent feature of the controller lies in its ability to achieve a controllable trade-off between the effectiveness of vibration suppression and control inputs. Moreover, the controller can effectively suppress vibrations even when cable forces are constrained. The underlying principle is that the vibration energy is consumed through the negative work done by the cable forces until the tip displacement is ultimately reduced to a small range. The stability of the controller is verified by the Lyapunov method. Simulations demonstrate the effectiveness of the proposed control scheme, while experimental validation on a prototype consisting of a six-meter-long flexible structure and a four-cable CDPR further supports these findings. - FNClarivate Analytics Web of Science
- VR1.0
- PTJ
- AUSun, HN
Tang, XQ
Ge, SS
- AFHaining Sun
Xiaoqiang Tang
Shuzhi Sam Ge
- TIControllable Trade-Off Between Performance and Constrained Input for Vibration Suppression in Flexible Space Structures
- SOIEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS
- LAEnglish
- DTArticle
- DEVibrations; Flexible Structures; Force; Aerospace Electronics; Vibration Control; Damping; Finite Element Analysis
- IDTRACKING; ATTITUDE; ROBOTS; BEAM
- ABAchieving vibration suppression of flexible space structures requires not only considering the vibration settling time, but also avoiding excessive control forces that might damage the structure. In this article, an active control scheme is designed to suppress undesired vibrations in flexible structures by exerting small cable forces. It employs a cable-driven parallel robot (CDPR) as an actuator. The prominent feature of the controller lies in its ability to achieve a controllable trade-off between the effectiveness of vibration suppression and control inputs. Moreover, the controller can effectively suppress vibrations even when cable forces are constrained. The underlying principle is that the vibration energy is consumed through the negative work done by the cable forces until the tip displacement is ultimately reduced to a small range. The stability of the controller is verified by the Lyapunov method. Simulations demonstrate the effectiveness of the proposed control scheme, while experimental validation on a prototype consisting of a six-meter-long flexible structure and a four-cable CDPR further supports these findings.
- C3Agency for Science Technology & Research (A*STAR); A*STAR - Singapore Institute of Manufacturing Technology (SIMTech); Tsinghua University; National University of Singapore; Institute for Functional Intelligent Materials (I-FIM)
- RPTang, XQ (corresponding author), Tsinghua Univ, Dept Mech Engn, State Key Lab Tribol, Beijing 100084, Peoples R China
- FXNo Statement Available
- NR52
- TC0
- Z90
- U10
- U20
- PUIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
- PIPISCATAWAY
- PA445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA
- SN0018-9251
- J9IEEE TRANS AEROSP ELECTRON SY
- JIIEEE Trans. Aerosp. Electron. Syst.
- PDJUN
- PY2024
- VL60
- BP3390
- EP3402
- DI10.1109/TAES.2024.3361434
- PG13
- WCEngineering, Aerospace; Engineering, Electrical & Electronic; Telecommunications
- SCEngineering; Telecommunications
- GAUF2O6
- UTWOS:001246582400007
- ER
- EF
|
Wang, Hao; Sun, Bin; Ge, Shuzhi Sam; Su, Jie; Jin, Ming Liang On non-von Neumann flexible neuromorphic vision sensors NPJ FLEXIBLE ELECTRONICS, 8 (1), 2024, DOI: 10.1038/s41528-024-00313-3. Abstract | BibTeX | Endnote @article{ISI:001215635300003,
title = {On non-von Neumann flexible neuromorphic vision sensors},
author = {Hao Wang and Bin Sun and Shuzhi Sam Ge and Jie Su and Ming Liang Jin},
doi = {10.1038/s41528-024-00313-3},
times_cited = {0},
year = {2024},
date = {2024-05-07},
journal = {NPJ FLEXIBLE ELECTRONICS},
volume = {8},
number = {1},
publisher = {NATURE PORTFOLIO},
address = {HEIDELBERGER PLATZ 3, BERLIN, 14197, GERMANY},
abstract = {The structure and mechanism of the human visual system contain rich treasures, and surprising effects can be achieved by simulating the human visual system. In this article, starting from the human visual system, we compare and discuss the discrepancies between the human visual system and traditional machine vision systems. Given the wide variety and large volume of visual information, the use of non-von Neumann structured, flexible neuromorphic vision sensors can effectively compensate for the limitations of traditional machine vision systems based on the von Neumann architecture. Firstly, this article addresses the emulation of retinal functionality and provides an overview of the principles and circuit implementation methods of non-von Neumann computing architectures. Secondly, in terms of mimicking the retinal surface structure, this article introduces the fabrication approach for flexible sensor arrays. Finally, this article analyzes the challenges currently faced by non-von Neumann flexible neuromorphic vision sensors and offers a perspective on their future development.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The structure and mechanism of the human visual system contain rich treasures, and surprising effects can be achieved by simulating the human visual system. In this article, starting from the human visual system, we compare and discuss the discrepancies between the human visual system and traditional machine vision systems. Given the wide variety and large volume of visual information, the use of non-von Neumann structured, flexible neuromorphic vision sensors can effectively compensate for the limitations of traditional machine vision systems based on the von Neumann architecture. Firstly, this article addresses the emulation of retinal functionality and provides an overview of the principles and circuit implementation methods of non-von Neumann computing architectures. Secondly, in terms of mimicking the retinal surface structure, this article introduces the fabrication approach for flexible sensor arrays. Finally, this article analyzes the challenges currently faced by non-von Neumann flexible neuromorphic vision sensors and offers a perspective on their future development. - FNClarivate Analytics Web of Science
- VR1.0
- PTJ
- AUWang, H
Sun, B
Ge, SS
Su, J
Jin, ML
- AFHao Wang
Bin Sun
Shuzhi Sam Ge
Jie Su
Ming Liang Jin
- TIOn non-von Neumann flexible neuromorphic vision sensors
- SONPJ FLEXIBLE ELECTRONICS
- LAEnglish
- DTArticle
- IDNEURAL-NETWORK; GANGLION-CELLS; ANALOG; PHOTODETECTOR; ULTRAFAST; MEMORY; PHOTOTRANSISTORS; PLASTICITY; MONOLAYER; FRAMEWORK
- ABThe structure and mechanism of the human visual system contain rich treasures, and surprising effects can be achieved by simulating the human visual system. In this article, starting from the human visual system, we compare and discuss the discrepancies between the human visual system and traditional machine vision systems. Given the wide variety and large volume of visual information, the use of non-von Neumann structured, flexible neuromorphic vision sensors can effectively compensate for the limitations of traditional machine vision systems based on the von Neumann architecture. Firstly, this article addresses the emulation of retinal functionality and provides an overview of the principles and circuit implementation methods of non-von Neumann computing architectures. Secondly, in terms of mimicking the retinal surface structure, this article introduces the fabrication approach for flexible sensor arrays. Finally, this article analyzes the challenges currently faced by non-von Neumann flexible neuromorphic vision sensors and offers a perspective on their future development.
- C1[Wang, Hao; Jin, Ming Liang] Qingdao Univ, Inst Future, Sch Automat, Shandong Key Lab Ind Control Technol, Qingdao 266071, Peoples R China.
[Sun, Bin; Su, Jie] Qingdao Univ, Coll Elect & Informat Engn, Qingdao 266071, Peoples R China. [Ge, Shuzhi Sam] Natl Univ Singapore, Dept Elect & Comp Engn, Singapore 117583, Singapore. [Ge, Shuzhi Sam] Natl Univ Singapore, Inst Funct Intelligent Mat, Singapore 117583, Singapore - C3Qingdao University; Qingdao University; National University of Singapore; Institute for Functional Intelligent Materials (I-FIM); National University of Singapore
- RPJin, ML (corresponding author), Qingdao Univ, Inst Future, Sch Automat, Shandong Key Lab Ind Control Technol, Qingdao 266071, Peoples R China; Su, J (corresponding author), Qingdao Univ, Coll Elect & Informat Engn, Qingdao 266071, Peoples R China; Ge, SS (corresponding author), Natl Univ Singapore, Dept Elect & Comp Engn, Singapore 117583, Singapore; Ge, SS (corresponding author), Natl Univ Singapore, Inst Funct Intelligent Mat, Singapore 117583, Singapore
- FUNational Natural Science Foundation of China (National Science Foundation of China) [201909099]; Young Taishan Scholars Program of Shandong Province [EDUNC-33-18-279-V12]; Ministry of Education, Singapore [52003134, 12374088]; National Natural Science Foundation of China
- FXThe work was financially supported by the Young Taishan Scholars Program of Shandong Province (grant nos. 201909099) to M.L. Jin, Ministry of Education, Singapore, under its Research Centre of Excellence award to the Institute for Functional Intelligent Materials (I-FIM, project No. EDUNC-33-18-279-V12) to S.S. Ge, and National Natural Science Foundation of China (grant nos. 52003134 and 12374088) to M.L. Jin and J. Su.
- NR164
- TC0
- Z90
- U17
- U27
- PUNATURE PORTFOLIO
- PIBERLIN
- PAHEIDELBERGER PLATZ 3, BERLIN, 14197, GERMANY
- J9NPJ FLEX ELECTRON
- JInpj Flex. Electron.
- PDMAY 7
- PY2024
- VL8
- DI10.1038/s41528-024-00313-3
- PG26
- WCEngineering, Electrical & Electronic; Materials Science, Multidisciplinary
- SCEngineering; Materials Science
- GAPQ8H5
- UTWOS:001215635300003
- ER
- EF
|
Zhang, Yuxiang; Liang, Xiaoling; Li, Dongyu; Ge, Shuzhi Sam; Gao, Bingzhao; Chen, Hong; Lee, Tong Heng Adaptive Safe Reinforcement Learning With Full-State Constraints and Constrained Adaptation for Autonomous Vehicles IEEE TRANSACTIONS ON CYBERNETICS, 54 (3), pp. 1907-1920, 2024, DOI: 10.1109/TCYB.2023.3283771. Abstract | BibTeX | Endnote @article{ISI:001203365100010,
title = {Adaptive Safe Reinforcement Learning With Full-State Constraints and Constrained Adaptation for Autonomous Vehicles},
author = {Yuxiang Zhang and Xiaoling Liang and Dongyu Li and Shuzhi Sam Ge and Bingzhao Gao and Hong Chen and Tong Heng Lee},
doi = {10.1109/TCYB.2023.3283771},
times_cited = {3},
issn = {2168-2267},
year = {2024},
date = {2024-03-01},
journal = {IEEE TRANSACTIONS ON CYBERNETICS},
volume = {54},
number = {3},
pages = {1907-1920},
publisher = {IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC},
address = {445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA},
abstract = {High-performance learning-based control for the typical safety-critical autonomous vehicles invariably requires that the full-state variables are constrained within the safety region even during the learning process. To solve this technically critical and challenging problem, this work proposes an adaptive safe reinforcement learning (RL) algorithm that invokes innovative safety-related RL methods with the consideration of constraining the full-state variables within the safety region with adaptation. These are developed toward assuring the attainment of the specified requirements on the full-state variables with two notable aspects. First, thus, an appropriately optimized backstepping technique and the asymmetric barrier Lyapunov function (BLF) methodology are used to establish the safe learning framework to ensure system full-state constraints requirements. More specifically, each subsystem's control and partial derivative of the value function are decomposed with asymmetric BLF-related items and an independent learning part. Then, the independent learning part is updated to solve the Hamilton-Jacobi-Bellman equation through an adaptive learning implementation to attain the desired performance in system control. Second, with further Lyapunov-based analysis, it is demonstrated that safety performance is effectively doubly assured via a methodology of a constrained adaptation algorithm during optimization (which incorporates the projection operator and can deal with the conflict between safety and optimization). Therefore, this algorithm optimizes system control and ensures that the full set of state variables involved is always constrained within the safety region during the whole learning process. Comparison simulations and ablation studies are carried out on motion control problems for autonomous vehicles, which have verified superior performance with smaller variance and better convergence performance under uncertain circumstances. The effectiveness of the safe performance of overall system control with the proposed method accordingly has been verified.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
High-performance learning-based control for the typical safety-critical autonomous vehicles invariably requires that the full-state variables are constrained within the safety region even during the learning process. To solve this technically critical and challenging problem, this work proposes an adaptive safe reinforcement learning (RL) algorithm that invokes innovative safety-related RL methods with the consideration of constraining the full-state variables within the safety region with adaptation. These are developed toward assuring the attainment of the specified requirements on the full-state variables with two notable aspects. First, thus, an appropriately optimized backstepping technique and the asymmetric barrier Lyapunov function (BLF) methodology are used to establish the safe learning framework to ensure system full-state constraints requirements. More specifically, each subsystem's control and partial derivative of the value function are decomposed with asymmetric BLF-related items and an independent learning part. Then, the independent learning part is updated to solve the Hamilton-Jacobi-Bellman equation through an adaptive learning implementation to attain the desired performance in system control. Second, with further Lyapunov-based analysis, it is demonstrated that safety performance is effectively doubly assured via a methodology of a constrained adaptation algorithm during optimization (which incorporates the projection operator and can deal with the conflict between safety and optimization). Therefore, this algorithm optimizes system control and ensures that the full set of state variables involved is always constrained within the safety region during the whole learning process. Comparison simulations and ablation studies are carried out on motion control problems for autonomous vehicles, which have verified superior performance with smaller variance and better convergence performance under uncertain circumstances. The effectiveness of the safe performance of overall system control with the proposed method accordingly has been verified. - FNClarivate Analytics Web of Science
- VR1.0
- PTJ
- AUZhang, YX
Liang, XL
Li, DY
Ge, SS
Gao, BZ
Chen, H
Lee, TH
- AFYuxiang Zhang
Xiaoling Liang
Dongyu Li
Shuzhi Sam Ge
Bingzhao Gao
Hong Chen
Tong Heng Lee
- TIAdaptive Safe Reinforcement Learning With Full-State Constraints and Constrained Adaptation for Autonomous Vehicles
- SOIEEE TRANSACTIONS ON CYBERNETICS
- LAEnglish
- DTArticle
- DEAdaptive Dynamic Programming (ADP); Autonomous Vehicles; Barrier Lyapunov Function (BLF); Safe Reinforcement Learning (RL)
- IDBARRIER LYAPUNOV FUNCTIONS; NONLINEAR-SYSTEMS; TRACKING CONTROL
- ABHigh-performance learning-based control for the typical safety-critical autonomous vehicles invariably requires that the full-state variables are constrained within the safety region even during the learning process. To solve this technically critical and challenging problem, this work proposes an adaptive safe reinforcement learning (RL) algorithm that invokes innovative safety-related RL methods with the consideration of constraining the full-state variables within the safety region with adaptation. These are developed toward assuring the attainment of the specified requirements on the full-state variables with two notable aspects. First, thus, an appropriately optimized backstepping technique and the asymmetric barrier Lyapunov function (BLF) methodology are used to establish the safe learning framework to ensure system full-state constraints requirements. More specifically, each subsystem's control and partial derivative of the value function are decomposed with asymmetric BLF-related items and an independent learning part. Then, the independent learning part is updated to solve the Hamilton-Jacobi-Bellman equation through an adaptive learning implementation to attain the desired performance in system control. Second, with further Lyapunov-based analysis, it is demonstrated that safety performance is effectively doubly assured via a methodology of a constrained adaptation algorithm during optimization (which incorporates the projection operator and can deal with the conflict between safety and optimization). Therefore, this algorithm optimizes system control and ensures that the full set of state variables involved is always constrained within the safety region during the whole learning process. Comparison simulations and ablation studies are carried out on motion control problems for autonomous vehicles, which have verified superior performance with smaller variance and better convergence performance under uncertain circumstances. The effectiveness of the safe performance of overall system control with the proposed method accordingly has been verified.
- C3National University of Singapore; Institute for Functional Intelligent Materials (I-FIM); National University of Singapore; National University of Singapore; Beihang University; Tongji University; Tongji University
- RPGe, SS (corresponding author), Natl Univ Singapore, Dept Elect & Comp Engn, Singapore 117583, Singapore; Ge, SS (corresponding author), Natl Univ Singapore, Inst Funct Intelligent Mat, Singapore 117583, Singapore; Gao, BZ (corresponding author), Tongji Univ, Clean Energy Automot Engn Ctr, Shanghai 201804, Peoples R China
- FXNo Statement Available
- NR48
- TC3
- Z93
- U126
- U247
- PUIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
- PIPISCATAWAY
- PA445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA
- SN2168-2267
- J9IEE TRANS CYBERN
- JIIEEE T. Cybern.
- PDMAR
- PY2024
- VL54
- BP1907
- EP1920
- DI10.1109/TCYB.2023.3283771
- PG14
- WCAutomation & Control Systems; Computer Science, Artificial Intelligence; Computer Science, Cybernetics
- SCAutomation & Control Systems; Computer Science
- UTWOS:001203365100010
- ER
- EF
|
Ji, Ruihang; Ge, Shuzhi Sam Event-Triggered Tunnel Prescribed Control for Nonlinear Systems IEEE TRANSACTIONS ON FUZZY SYSTEMS, 32 (1), pp. 90-101, 2024, DOI: 10.1109/TFUZZ.2023.3290934. Abstract | BibTeX | Endnote @article{ISI:001136745800012,
title = {Event-Triggered Tunnel Prescribed Control for Nonlinear Systems},
author = {Ruihang Ji and Shuzhi Sam Ge},
doi = {10.1109/TFUZZ.2023.3290934},
times_cited = {3},
issn = {1063-6706},
year = {2024},
date = {2024-01-01},
journal = {IEEE TRANSACTIONS ON FUZZY SYSTEMS},
volume = {32},
number = {1},
pages = {90-101},
publisher = {IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC},
address = {445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA},
abstract = {This article studies an event-triggered tunnel prescribed control (TPC) for uncertain nonlinear systems under any initial condition. A more general entry capture problem (ECP) is introduced, where the tunnel prescribed performance is satisfied after a certain period of system operation, as opposed to starting from the beginning, thereby, making the control design complex yet challenging. In this case, the normally employed prescribed performance control becomes invalid due to the singularity problem arising from the initial condition violation. An error self-tuning function is proposed to provide a unified approach for handling different initial conditions, which can be extended to other methods. In order to deal with unknown control directions, an orientation function is employed in lieu of Nussbaum-type function, by which the initial control input is always equal to zero avoiding a large initial value. We then develop an event-triggered mechanism from an encoding-decoding viewpoint, by which only 1-bit string, either 1 or 0, is required for each communication between control and actuator. In this way, such event-triggered mechanism can further reduce communication burden and improve communication security at the same time. The developed event-triggered TPC provides an effective solution for the underlying ECP and exhibits low-complexity level since no additional filters or time derivatives of virtual control inputs are required in the control design. Finally, comparative simulations are conducted to illustrate the abovementioned theoretical finding.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This article studies an event-triggered tunnel prescribed control (TPC) for uncertain nonlinear systems under any initial condition. A more general entry capture problem (ECP) is introduced, where the tunnel prescribed performance is satisfied after a certain period of system operation, as opposed to starting from the beginning, thereby, making the control design complex yet challenging. In this case, the normally employed prescribed performance control becomes invalid due to the singularity problem arising from the initial condition violation. An error self-tuning function is proposed to provide a unified approach for handling different initial conditions, which can be extended to other methods. In order to deal with unknown control directions, an orientation function is employed in lieu of Nussbaum-type function, by which the initial control input is always equal to zero avoiding a large initial value. We then develop an event-triggered mechanism from an encoding-decoding viewpoint, by which only 1-bit string, either 1 or 0, is required for each communication between control and actuator. In this way, such event-triggered mechanism can further reduce communication burden and improve communication security at the same time. The developed event-triggered TPC provides an effective solution for the underlying ECP and exhibits low-complexity level since no additional filters or time derivatives of virtual control inputs are required in the control design. Finally, comparative simulations are conducted to illustrate the abovementioned theoretical finding. - FNClarivate Analytics Web of Science
- VR1.0
- PTJ
- AUJi, RH
Ge, SS
- AFRuihang Ji
Shuzhi Sam Ge
- TIEvent-Triggered Tunnel Prescribed Control for Nonlinear Systems
- SOIEEE TRANSACTIONS ON FUZZY SYSTEMS
- LAEnglish
- DTArticle
- DEEvent-triggered Control; Global Property; Nonlinear Systems; Prescribed Performance Control (PPC)
- IDPERFORMANCE CONTROL; ADAPTIVE-CONTROL; TRACKING CONTROL; STATE
- ABThis article studies an event-triggered tunnel prescribed control (TPC) for uncertain nonlinear systems under any initial condition. A more general entry capture problem (ECP) is introduced, where the tunnel prescribed performance is satisfied after a certain period of system operation, as opposed to starting from the beginning, thereby, making the control design complex yet challenging. In this case, the normally employed prescribed performance control becomes invalid due to the singularity problem arising from the initial condition violation. An error self-tuning function is proposed to provide a unified approach for handling different initial conditions, which can be extended to other methods. In order to deal with unknown control directions, an orientation function is employed in lieu of Nussbaum-type function, by which the initial control input is always equal to zero avoiding a large initial value. We then develop an event-triggered mechanism from an encoding-decoding viewpoint, by which only 1-bit string, either 1 or 0, is required for each communication between control and actuator. In this way, such event-triggered mechanism can further reduce communication burden and improve communication security at the same time. The developed event-triggered TPC provides an effective solution for the underlying ECP and exhibits low-complexity level since no additional filters or time derivatives of virtual control inputs are required in the control design. Finally, comparative simulations are conducted to illustrate the abovementioned theoretical finding.
- C3National University of Singapore; Institute for Functional Intelligent Materials (I-FIM); National University of Singapore
- RPGe, SS (corresponding author), Natl Univ Singapore, Inst Funct Intelligent Mat, Singapore 117544, Singapore
- FXNo Statement Available
- NR49
- TC3
- Z93
- U120
- U220
- PUIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
- PIPISCATAWAY
- PA445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA
- SN1063-6706
- J9IEEE TRANS FUZZY SYST
- JIIEEE Trans. Fuzzy Syst.
- PDJAN
- PY2024
- VL32
- BP90
- EP101
- DI10.1109/TFUZZ.2023.3290934
- PG12
- WCComputer Science, Artificial Intelligence; Engineering, Electrical & Electronic
- SCComputer Science; Engineering
- GAEC6D2
- UTWOS:001136745800012
- ER
- EF
|
2023
|
Yang, Zeyuan; Xu, Xiaohu; Li, Jie; Zhu, Dahu; Yan, Sijie; Ge, Shuzhi Sam; Ding, Han Knowledge-wrapping method for prediction and evaluation of material removal behavior in robotic belt grinding MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 208 , 2023, DOI: 10.1016/j.ymssp.2023.110914. Abstract | BibTeX | Endnote @article{ISI:001135283600001,
title = {Knowledge-wrapping method for prediction and evaluation of material removal behavior in robotic belt grinding},
author = {Zeyuan Yang and Xiaohu Xu and Jie Li and Dahu Zhu and Sijie Yan and Shuzhi Sam Ge and Han Ding},
doi = {10.1016/j.ymssp.2023.110914},
times_cited = {2},
issn = {0888-3270},
year = {2023},
date = {2023-12-02},
journal = {MECHANICAL SYSTEMS AND SIGNAL PROCESSING},
volume = {208},
publisher = {ACADEMIC PRESS LTD- ELSEVIER SCIENCE LTD},
address = {24-28 OVAL RD, LONDON NW1 7DX, ENGLAND},
abstract = {Reliable prediction and evaluation of material removal (MR) is a continuing pursuit in the grinding process. Mechanistic and empirical MR models always suffer from inaccuracies and restricted applicability, whereas data-driven approaches remain deficient in sample dependence, generalization, and physical interpretability. This motivates us to develop a knowledge-wrapping method (KWM) to predict and characterize the material removal behavior in robotic belt grinding. A physical material removal profile model (PHY) that suits the robotic belt grinding is first presented by appealing to the Archard law and Hertzian theory. Next, the knowledge-wrapping matrix is designed to wrap the physical constraints into a form of design matrix by transforming the mechanistic and empirical models into a linear system. The knowledge-wrapping matrix and the experimental data are then connected via a likelihood function with the consideration of measuring noise, yielding a hybrid-driven learning process that preserves interpretability from PHY. Comparative experiments and MR mechanism interpretations are finally presented to demonstrate the effectiveness and superiority of the proposed method.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Reliable prediction and evaluation of material removal (MR) is a continuing pursuit in the grinding process. Mechanistic and empirical MR models always suffer from inaccuracies and restricted applicability, whereas data-driven approaches remain deficient in sample dependence, generalization, and physical interpretability. This motivates us to develop a knowledge-wrapping method (KWM) to predict and characterize the material removal behavior in robotic belt grinding. A physical material removal profile model (PHY) that suits the robotic belt grinding is first presented by appealing to the Archard law and Hertzian theory. Next, the knowledge-wrapping matrix is designed to wrap the physical constraints into a form of design matrix by transforming the mechanistic and empirical models into a linear system. The knowledge-wrapping matrix and the experimental data are then connected via a likelihood function with the consideration of measuring noise, yielding a hybrid-driven learning process that preserves interpretability from PHY. Comparative experiments and MR mechanism interpretations are finally presented to demonstrate the effectiveness and superiority of the proposed method. - FNClarivate Analytics Web of Science
- VR1.0
- PTJ
- AUYang, ZY
Xu, XH
Li, J
Zhu, DH
Yan, SJ
Ge, SS
Ding, H
- AFZeyuan Yang
Xiaohu Xu
Jie Li
Dahu Zhu
Sijie Yan
Shuzhi Sam Ge
Han Ding
- TIKnowledge-wrapping method for prediction and evaluation of material removal behavior in robotic belt grinding
- SOMECHANICAL SYSTEMS AND SIGNAL PROCESSING
- LAEnglish
- DTArticle
- DERobotic Grinding; Material Removal; Physics-data Driven; Knowledge Learning
- IDMODEL; SIZE; SIMULATION; MECHANICS; DEPTH
- ABReliable prediction and evaluation of material removal (MR) is a continuing pursuit in the grinding process. Mechanistic and empirical MR models always suffer from inaccuracies and restricted applicability, whereas data-driven approaches remain deficient in sample dependence, generalization, and physical interpretability. This motivates us to develop a knowledge-wrapping method (KWM) to predict and characterize the material removal behavior in robotic belt grinding. A physical material removal profile model (PHY) that suits the robotic belt grinding is first presented by appealing to the Archard law and Hertzian theory. Next, the knowledge-wrapping matrix is designed to wrap the physical constraints into a form of design matrix by transforming the mechanistic and empirical models into a linear system. The knowledge-wrapping matrix and the experimental data are then connected via a likelihood function with the consideration of measuring noise, yielding a hybrid-driven learning process that preserves interpretability from PHY. Comparative experiments and MR mechanism interpretations are finally presented to demonstrate the effectiveness and superiority of the proposed method.
- C1[Yang, Zeyuan; Xu, Xiaohu; Li, Jie; Yan, Sijie; Ding, Han] Huazhong Univ Sci & Technol, State Key Lab Intelligent Mfg Equipment & Technol, Wuhan 430074, Peoples R China.
[Xu, Xiaohu] Wuhan Univ, Inst Technol Sci, Wuhan 430072, Peoples R China. [Zhu, Dahu] Wuhan Univ Technol, Hubei Key Lab Adv Technol Automot Components, Wuhan 430070, Peoples R China. [Yang, Zeyuan; Ge, Shuzhi Sam] Natl Univ Singapore, Dept Elect & Comp Engn, Singapore 117576, Singapore. [Yang, Zeyuan; Li, Jie] HUST Wuxi Res Inst, Wuxi 214174, Peoples R China - C3Huazhong University of Science & Technology; Wuhan University; Wuhan University of Technology; National University of Singapore
- RPYan, SJ (corresponding author), Huazhong Univ Sci & Technol, State Key Lab Intelligent Mfg Equipment & Technol, Wuhan 430074, Peoples R China
- FUNational Key Research and Development Program of China [2019YFA0706703]; Fundamental Research Funds for the Central Universities, China [2042023kf0114]; National Nature Science Foundation of China [52075204, 51975443, 52105514]; Research Centre of Excellence award [EDUNC-33-18-279-V12]; China Scholarship Council [202106160036]; Ministry of Education, Singapore;
- FXThis research is supported by the National Key Research and Development Program of China (No. 2019YFA0706703) , the Fundamental Research Funds for the Central Universities, China (No. 2042023kf0114) , the National Nature Science Foundation of China (Nos. 52075204, 51975443, 52105514) , the Research Centre of Excellence award to the Institute for Functional Intelligent Materials, Ministry of Education, Singapore (No. EDUNC-33-18-279-V12) , and the China Scholarship Council (No. 202106160036) .
- NR55
- TC2
- Z92
- U19
- U29
- PUACADEMIC PRESS LTD- ELSEVIER SCIENCE LTD
- PILONDON
- PA24-28 OVAL RD, LONDON NW1 7DX, ENGLAND
- SN0888-3270
- J9MECH SYST SIGNAL PROCESS
- JIMech. Syst. Signal Proc.
- PDFEB 15
- PY2024
- VL208
- DI10.1016/j.ymssp.2023.110914
- PG16
- WCEngineering, Mechanical
- SCEngineering
- GADX1G6
- UTWOS:001135283600001
- ER
- EF
|