Pub Date : 2023-10-18DOI: 10.1177/01423312231200058
Zhangyu Zhou, Feng Xu, Yuchen Dai
An integral sliding mode observer-based fixed-time multi-faults estimation method is proposed to solve the problem of fast fault estimation for a class of nonlinear systems with multiple types of faults simultaneously. First, a class of nonlinear systems with actuator and sensor faults is considered, and the original system is transformed into two subsystems through coordinate transformation to decouple the actuator and sensor faults. Then, intermediate variables are introduced to transform sensor faults into actuator faults of the augmented systems. For the transformed system, a fixed-time fault observer based on integral sliding mode is designed to realize the fast estimation of unknown terms, and the fixed-time convergence performance is proved. Moreover, based on the output of the fault observer, the actuator and sensor faults in the original system are reconstructed. Finally, the fault model of the permanent magnet synchronous motor is established in the hardware-in-the-loop platform, and the effectiveness of the designed fault estimation method is verified.
{"title":"Fixed-time sliding mode observer-based multi-faults estimation method and its application to PMSM drive system","authors":"Zhangyu Zhou, Feng Xu, Yuchen Dai","doi":"10.1177/01423312231200058","DOIUrl":"https://doi.org/10.1177/01423312231200058","url":null,"abstract":"An integral sliding mode observer-based fixed-time multi-faults estimation method is proposed to solve the problem of fast fault estimation for a class of nonlinear systems with multiple types of faults simultaneously. First, a class of nonlinear systems with actuator and sensor faults is considered, and the original system is transformed into two subsystems through coordinate transformation to decouple the actuator and sensor faults. Then, intermediate variables are introduced to transform sensor faults into actuator faults of the augmented systems. For the transformed system, a fixed-time fault observer based on integral sliding mode is designed to realize the fast estimation of unknown terms, and the fixed-time convergence performance is proved. Moreover, based on the output of the fault observer, the actuator and sensor faults in the original system are reconstructed. Finally, the fault model of the permanent magnet synchronous motor is established in the hardware-in-the-loop platform, and the effectiveness of the designed fault estimation method is verified.","PeriodicalId":49426,"journal":{"name":"Transactions of the Institute of Measurement and Control","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135824078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-18DOI: 10.1177/01423312231200340
Ding Duan, Yuling Li, Hongbing Li
Safety, stability, and intuitive control are the three most basic requirements for the design of a physical human–robot cooperation (HRC) system. High transparency is one of the main design objectives for dynamically coupled HRC systems. However, determinants of system transparency and its influence on system performance are unachievable with the current methods available. This paper presents an overview of admittance control as a method of physical interaction control between robots and human operators, in particular on the analysis of the influencing factors to the computational transparency index. The influence of frequency, sampling point number, and cutoff frequency of the weighting function on the transparency of the admittance controller were analyzed and experimentally verified. The controller was implemented on a four degree-of-freedom interactive manipulator to verify system transparency and stability. The experimental results validate the proposed framework.
{"title":"Factors affecting trust in the transparency index for stable and intuitive physical human–robot cooperation","authors":"Ding Duan, Yuling Li, Hongbing Li","doi":"10.1177/01423312231200340","DOIUrl":"https://doi.org/10.1177/01423312231200340","url":null,"abstract":"Safety, stability, and intuitive control are the three most basic requirements for the design of a physical human–robot cooperation (HRC) system. High transparency is one of the main design objectives for dynamically coupled HRC systems. However, determinants of system transparency and its influence on system performance are unachievable with the current methods available. This paper presents an overview of admittance control as a method of physical interaction control between robots and human operators, in particular on the analysis of the influencing factors to the computational transparency index. The influence of frequency, sampling point number, and cutoff frequency of the weighting function on the transparency of the admittance controller were analyzed and experimentally verified. The controller was implemented on a four degree-of-freedom interactive manipulator to verify system transparency and stability. The experimental results validate the proposed framework.","PeriodicalId":49426,"journal":{"name":"Transactions of the Institute of Measurement and Control","volume":"65 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135884023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-17DOI: 10.1177/01423312231198953
Xiufang Wang, Shuang Sun, Chunlei Jiang, Hongbo Bi, Wendi Yan
Compound fault signals interfere with one another, resulting in an inconspicuous feature extraction that requires sophisticated signal processing techniques and expert experience. However, good online diagnostic methods are not available to carry out this process. This paper proposes a method based on Residual Connection and Squeeze-and-Excitation Unet (RSEUnet) and one-dimensional convolutional neural network (1DCNN). The process includes fault separation and diagnosis. First, the feature extraction module of the RSEUnet network introduces an attention mechanism and a residual connection that adaptively assigns various weights to different channels. This model is used to train the maps of the fault signal after time-frequency transformation. Ideal binary masks with excellent performance are the training targets to complete the intelligent separation of compound faults. Second, the 1DCNN is used as a feature learning model to efficiently learn the features of single faults from time-domain signals. An embedded system consisting of a Jetson Nano and a signal acquisition circuit is then built to perform online diagnosis. The test is carried out on the fault experimental platform. Results show that the method has an accuracy of 99.71%, making it highly suitable for the diagnosis of bearing compound faults.
{"title":"Compound fault separation and diagnosis method for online rolling bearings based on RSEUnet and 1DCNN","authors":"Xiufang Wang, Shuang Sun, Chunlei Jiang, Hongbo Bi, Wendi Yan","doi":"10.1177/01423312231198953","DOIUrl":"https://doi.org/10.1177/01423312231198953","url":null,"abstract":"Compound fault signals interfere with one another, resulting in an inconspicuous feature extraction that requires sophisticated signal processing techniques and expert experience. However, good online diagnostic methods are not available to carry out this process. This paper proposes a method based on Residual Connection and Squeeze-and-Excitation Unet (RSEUnet) and one-dimensional convolutional neural network (1DCNN). The process includes fault separation and diagnosis. First, the feature extraction module of the RSEUnet network introduces an attention mechanism and a residual connection that adaptively assigns various weights to different channels. This model is used to train the maps of the fault signal after time-frequency transformation. Ideal binary masks with excellent performance are the training targets to complete the intelligent separation of compound faults. Second, the 1DCNN is used as a feature learning model to efficiently learn the features of single faults from time-domain signals. An embedded system consisting of a Jetson Nano and a signal acquisition circuit is then built to perform online diagnosis. The test is carried out on the fault experimental platform. Results show that the method has an accuracy of 99.71%, making it highly suitable for the diagnosis of bearing compound faults.","PeriodicalId":49426,"journal":{"name":"Transactions of the Institute of Measurement and Control","volume":"173 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136032935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-17DOI: 10.1177/01423312231199807
Liu Yang, Chunhui Li, Qingtao Wang, Shuxian Shi, Mengru Yang
Aiming at the problem that the dynamic window approach (DWA) has low planning efficiency and it is difficult to avoid fast-moving obstacles, an improved adaptive DWA based on risk degree function is proposed in this paper. The risk function is designed and introduced into the evaluation function of the traditional DWA to evaluate the risk of collision between dynamic obstacles and the robot, so that the robot can effectively avoid faster obstacles. Then, according to the fuzzy control principle, the adaptive weight coefficient is designed to improve the evaluation function, so that the mobile robot can move to the target point more efficiently. The simulation results show that compared with the traditional DWA, the adaptive DWA based on risk degree function reduces the time of completing the planning task by about 8%, and the path length after the completion of the planning by about 8%, it indicates that the improved algorithm has higher efficiency. After 50 repeated experiments, using the adaptive DWA based on risk degree function successfully avoids all obstacles to complete the planning task, which shows that this algorithm has higher security.
{"title":"Adaptive dynamic windowing approach based on risk degree function","authors":"Liu Yang, Chunhui Li, Qingtao Wang, Shuxian Shi, Mengru Yang","doi":"10.1177/01423312231199807","DOIUrl":"https://doi.org/10.1177/01423312231199807","url":null,"abstract":"Aiming at the problem that the dynamic window approach (DWA) has low planning efficiency and it is difficult to avoid fast-moving obstacles, an improved adaptive DWA based on risk degree function is proposed in this paper. The risk function is designed and introduced into the evaluation function of the traditional DWA to evaluate the risk of collision between dynamic obstacles and the robot, so that the robot can effectively avoid faster obstacles. Then, according to the fuzzy control principle, the adaptive weight coefficient is designed to improve the evaluation function, so that the mobile robot can move to the target point more efficiently. The simulation results show that compared with the traditional DWA, the adaptive DWA based on risk degree function reduces the time of completing the planning task by about 8%, and the path length after the completion of the planning by about 8%, it indicates that the improved algorithm has higher efficiency. After 50 repeated experiments, using the adaptive DWA based on risk degree function successfully avoids all obstacles to complete the planning task, which shows that this algorithm has higher security.","PeriodicalId":49426,"journal":{"name":"Transactions of the Institute of Measurement and Control","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136032638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-16DOI: 10.1177/01423312231198920
Oussama Bey, Mohamed Chemachema
In this paper, a control-error-based decentralized neural network (NN) direct adaptive controller is presented for uncertain interconnected nonlinear systems, in strict-feedback form, subject to input saturation and external disturbances with unavailable states for measurement. Different from the existing results in the literature, the proposed approach is based on the control error instead of the tracking error resulting in a separation-like principle. Furthermore, the explosion of complexity due to back-stepping recursive design is completely avoided along with discarding all restrictive assumptions imposed on the unmatched interconnections. Actually, NNs are used to approximate the unknown ideal control laws, and auxiliary control terms are appended to deal with approximation errors and enhance the stability of the closed-loop system. Besides, fuzzy inference systems are introduced to estimate the unknown control errors, leading to simplified derivation of adaptive laws. Thanks to the strictly positive real (SPR) property, the tracking errors are proved to converge asymptotically to zero using Lyapunov theory, which is superior to bounded stability results usually found in the literature. Simulation results show the effectiveness of the proposed approach.
{"title":"Control-error-based output-feedback adaptive decentralized neural network controller for interconnected uncertain strict-feedback nonlinear systems with input saturation","authors":"Oussama Bey, Mohamed Chemachema","doi":"10.1177/01423312231198920","DOIUrl":"https://doi.org/10.1177/01423312231198920","url":null,"abstract":"In this paper, a control-error-based decentralized neural network (NN) direct adaptive controller is presented for uncertain interconnected nonlinear systems, in strict-feedback form, subject to input saturation and external disturbances with unavailable states for measurement. Different from the existing results in the literature, the proposed approach is based on the control error instead of the tracking error resulting in a separation-like principle. Furthermore, the explosion of complexity due to back-stepping recursive design is completely avoided along with discarding all restrictive assumptions imposed on the unmatched interconnections. Actually, NNs are used to approximate the unknown ideal control laws, and auxiliary control terms are appended to deal with approximation errors and enhance the stability of the closed-loop system. Besides, fuzzy inference systems are introduced to estimate the unknown control errors, leading to simplified derivation of adaptive laws. Thanks to the strictly positive real (SPR) property, the tracking errors are proved to converge asymptotically to zero using Lyapunov theory, which is superior to bounded stability results usually found in the literature. Simulation results show the effectiveness of the proposed approach.","PeriodicalId":49426,"journal":{"name":"Transactions of the Institute of Measurement and Control","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136113503","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-16DOI: 10.1177/01423312231200040
Huifang Min, Shang Shi, Hongyan Feng
This paper investigates the adaptive control for a class of constrained stochastic nonlinear systems with parametric uncertainty and input saturation. Based on a novel radial basis function neural networks (RBF NNs), the nonlinearities are tackled without the prior knowledge of NN nodes and weights. The approximate coordinate coordination is combined with an auxiliary system to attenuate the effects generated by input saturation. Then, an opportune backstepping design procedure is presented using the barrier Lyapunov function (BLF) and RBF NN. Based on this design procedure, an adaptive state–feedback controller is constructed, which makes the closed-loop system semi-globally uniformly ultimately bounded, the tracking error bounded in a compact set of the origin, and the full-states not violated. Finally, a stochastic single-link robot arm system is simulated to demonstrate the effectiveness of the proposed scheme.
{"title":"Adaptive stabilization of constrained stochastic nonlinear systems with input saturation: A combined BLF and NN approach","authors":"Huifang Min, Shang Shi, Hongyan Feng","doi":"10.1177/01423312231200040","DOIUrl":"https://doi.org/10.1177/01423312231200040","url":null,"abstract":"This paper investigates the adaptive control for a class of constrained stochastic nonlinear systems with parametric uncertainty and input saturation. Based on a novel radial basis function neural networks (RBF NNs), the nonlinearities are tackled without the prior knowledge of NN nodes and weights. The approximate coordinate coordination is combined with an auxiliary system to attenuate the effects generated by input saturation. Then, an opportune backstepping design procedure is presented using the barrier Lyapunov function (BLF) and RBF NN. Based on this design procedure, an adaptive state–feedback controller is constructed, which makes the closed-loop system semi-globally uniformly ultimately bounded, the tracking error bounded in a compact set of the origin, and the full-states not violated. Finally, a stochastic single-link robot arm system is simulated to demonstrate the effectiveness of the proposed scheme.","PeriodicalId":49426,"journal":{"name":"Transactions of the Institute of Measurement and Control","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136114583","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-14DOI: 10.1177/01423312231198950
Zhenyu Chang, Hongjing Liang, Liang Cao
This paper considers the security bipartite synchronization problem of nonlinear multi-agent systems (MASs) under the impact of denial-of-service (DoS) attacks. Notice that malicious attackers may block the neighbor information transmission among agents, which will reduce the efficiency of the distributed control strategy and even paralyze the entire system. In order to eliminate the impact of DoS attacks on the stability of MASs, this paper represents the first attempt to solve the security bipartite synchronization problem with the help of a distributed leader state observer. An observer that can obtain the leader state estimation is established, and the control strategy constructed based on the observer can eliminate the impact of DoS attacks. Theoretical analysis shows that the control strategy proposed in this paper can realize the security bipartite synchronization of the system under the impact of DoS attacks. The validity of the proposed results is verified by several presented simulation results.
{"title":"Security bipartite synchronization of MASs resilient to DoS attacks","authors":"Zhenyu Chang, Hongjing Liang, Liang Cao","doi":"10.1177/01423312231198950","DOIUrl":"https://doi.org/10.1177/01423312231198950","url":null,"abstract":"This paper considers the security bipartite synchronization problem of nonlinear multi-agent systems (MASs) under the impact of denial-of-service (DoS) attacks. Notice that malicious attackers may block the neighbor information transmission among agents, which will reduce the efficiency of the distributed control strategy and even paralyze the entire system. In order to eliminate the impact of DoS attacks on the stability of MASs, this paper represents the first attempt to solve the security bipartite synchronization problem with the help of a distributed leader state observer. An observer that can obtain the leader state estimation is established, and the control strategy constructed based on the observer can eliminate the impact of DoS attacks. Theoretical analysis shows that the control strategy proposed in this paper can realize the security bipartite synchronization of the system under the impact of DoS attacks. The validity of the proposed results is verified by several presented simulation results.","PeriodicalId":49426,"journal":{"name":"Transactions of the Institute of Measurement and Control","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135803879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-14DOI: 10.1177/01423312231198559
Meng-Jie Hu, Xinchen Shi, Zhuoqi Sun
In this study, the positive stability and [Formula: see text]-gain characterization problem are investigated for positive sampled-data systems. By constructing a new sampling-period-dependent copositive Lyapunov functional, and introducing the free weight matrix method to remove the single integral term, sufficient conditions for the asynchronous stability of the positive sampled-data system are established. Furthermore, system disturbances that occur frequently in engineering are taken into account. Conditions for robust stability of positive sampled-data systems that satisfy [Formula: see text]-gain performance are developed in the form of the linear programming technology. A numerical example and a practical simulation based on a communication network model are provided to illustrate the rationality and application of the theoretical results.
{"title":"Stability analysis and L1-gain characterization of positive sampled-data systems","authors":"Meng-Jie Hu, Xinchen Shi, Zhuoqi Sun","doi":"10.1177/01423312231198559","DOIUrl":"https://doi.org/10.1177/01423312231198559","url":null,"abstract":"In this study, the positive stability and [Formula: see text]-gain characterization problem are investigated for positive sampled-data systems. By constructing a new sampling-period-dependent copositive Lyapunov functional, and introducing the free weight matrix method to remove the single integral term, sufficient conditions for the asynchronous stability of the positive sampled-data system are established. Furthermore, system disturbances that occur frequently in engineering are taken into account. Conditions for robust stability of positive sampled-data systems that satisfy [Formula: see text]-gain performance are developed in the form of the linear programming technology. A numerical example and a practical simulation based on a communication network model are provided to illustrate the rationality and application of the theoretical results.","PeriodicalId":49426,"journal":{"name":"Transactions of the Institute of Measurement and Control","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135804146","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-14DOI: 10.1177/01423312231198567
Feng Gao, Yuan Zheng, Yan Li, Wenqiang Li
Owing to the lack of prior knowledge, the accurate reconstruction of surfaces with high uncertainty is dependent on the reasonable real-time selection of the next best point (NBP) during the sampling process. In this study, a new informative criterion called the MaxCWVar weighting shape effect is proposed for NBP selection. The responses to the geometric features of the candidate locations are predicted by a back propagation neural network (BPNN), which is then used in combination with the jackknife method to estimate the candidate uncertainty. The blade cross-section sampling case is considered to validate the flexibility and effectiveness of the proposed method. A comparison with other adaptive sampling strategies shows that BPNN-based response prediction is well-suited for allocating sample points. In contrast to other NBP selection criteria, the sample point distribution recommended by the MaxCWVar criterion is preferable as it improves the reconstruction accuracy and modeling efficiency. This study promotes the exploration of metrological methods for the fast and intelligent reconstruction of complex surfaces with high uncertainty.
{"title":"A back propagation neural network-based adaptive sampling strategy for uncertainty surfaces","authors":"Feng Gao, Yuan Zheng, Yan Li, Wenqiang Li","doi":"10.1177/01423312231198567","DOIUrl":"https://doi.org/10.1177/01423312231198567","url":null,"abstract":"Owing to the lack of prior knowledge, the accurate reconstruction of surfaces with high uncertainty is dependent on the reasonable real-time selection of the next best point (NBP) during the sampling process. In this study, a new informative criterion called the MaxCWVar weighting shape effect is proposed for NBP selection. The responses to the geometric features of the candidate locations are predicted by a back propagation neural network (BPNN), which is then used in combination with the jackknife method to estimate the candidate uncertainty. The blade cross-section sampling case is considered to validate the flexibility and effectiveness of the proposed method. A comparison with other adaptive sampling strategies shows that BPNN-based response prediction is well-suited for allocating sample points. In contrast to other NBP selection criteria, the sample point distribution recommended by the MaxCWVar criterion is preferable as it improves the reconstruction accuracy and modeling efficiency. This study promotes the exploration of metrological methods for the fast and intelligent reconstruction of complex surfaces with high uncertainty.","PeriodicalId":49426,"journal":{"name":"Transactions of the Institute of Measurement and Control","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135803886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-13DOI: 10.1177/01423312231197854
Zhaokang Xie, Xiaoci Huang, Suyun Luo, Ruoping Zhang, Fang Ma
To solve the problem of trajectory tracking lateral control in autonomous driving technology, a model predictive control (MPC) controller trajectory tracking lateral control method combined with a deep deterministic policy gradient algorithm (DDPG) is proposed in this paper. This method inputs the real-time state of the vehicle into DDPG to achieve real-time automatic optimization of the prediction time domain and control time domain parameters of the MPC controller, and then affects the specific performance of the MPC controller in trajectory tracking lateral control. Specifically, the state space, action space, and reward function of DDPG are defined, and the automatic driving trajectory tracking lateral controller is designed in combination with the vehicle dynamics model. To reduce the exploration space of DDPG and improve the training efficiency of the entire model, the technique of advantage-disadvantage experience separation and extraction is introduced. Finally, the proposed method was trained and verified in various scenarios, and compared with two other lateral control methods for autonomous driving. The results showed that the learning and training time of the trajectory tracking lateral control method based on DDPG-MPC was shorter than that of the DDPG-based method, and the evaluation indicators in the trajectory tracking control process were better than those of the DDPG-based method and original MPC-based method.
{"title":"A model predictive control trajectory tracking lateral controller for autonomous vehicles combined with deep deterministic policy gradient","authors":"Zhaokang Xie, Xiaoci Huang, Suyun Luo, Ruoping Zhang, Fang Ma","doi":"10.1177/01423312231197854","DOIUrl":"https://doi.org/10.1177/01423312231197854","url":null,"abstract":"To solve the problem of trajectory tracking lateral control in autonomous driving technology, a model predictive control (MPC) controller trajectory tracking lateral control method combined with a deep deterministic policy gradient algorithm (DDPG) is proposed in this paper. This method inputs the real-time state of the vehicle into DDPG to achieve real-time automatic optimization of the prediction time domain and control time domain parameters of the MPC controller, and then affects the specific performance of the MPC controller in trajectory tracking lateral control. Specifically, the state space, action space, and reward function of DDPG are defined, and the automatic driving trajectory tracking lateral controller is designed in combination with the vehicle dynamics model. To reduce the exploration space of DDPG and improve the training efficiency of the entire model, the technique of advantage-disadvantage experience separation and extraction is introduced. Finally, the proposed method was trained and verified in various scenarios, and compared with two other lateral control methods for autonomous driving. The results showed that the learning and training time of the trajectory tracking lateral control method based on DDPG-MPC was shorter than that of the DDPG-based method, and the evaluation indicators in the trajectory tracking control process were better than those of the DDPG-based method and original MPC-based method.","PeriodicalId":49426,"journal":{"name":"Transactions of the Institute of Measurement and Control","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135859050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","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：