المؤلفون: Yafeng Li, Jin Lei Wang, Aimin An

المصدر: IET Cyber-systems and Robotics (2020)

مصطلحات موضوعية: Computer Networks and Communications, Computer science, learning law, media_common.quotation_subject, Computer Science::Neural and Evolutionary Computation, MathematicsofComputing_NUMERICALANALYSIS, medical robotics, Dynamic modelling, variable parameters particle swarm optimisation, dynamic model, Inertia, patient rehabilitation, lcsh:QA75.5-76.95, dynamic modelling, Artificial Intelligence, Control theory, particle swarm optimisation, media_common, algorithm changes, Estimation theory, lcsh:Q300-390, Particle swarm optimization, uncertain parameters, dynamic parameter identification method, Function (mathematics), variable parameter particle swarm optimisation, upper-limb rehabilitation robot system, upper-limb rehabilitation robots, Human-Computer Interaction, Variable (computer science), Identification (information), Computational Theory and Mathematics, Hardware and Architecture, fixed-parameter, Control system, inertia parameter, lcsh:Electronic computers. Computer science, parameter estimation, lcsh:Cybernetics, identification accuracy, basic pso algorithm, Information Systems

الوصف: To solve the problem of uncertain parameters in dynamic modelling of upper-limb rehabilitation robots, a dynamic parameter identification method based on variable parameters particle swarm optimisation (PSO) is developed. Based on the dynamic model of the system, the algorithm changes the inertia parameter and learning law of the basic PSO algorithm from the fixed-parameter to the function that changes with the number of iterations. It solves the problems of small search space in the early stage and slow convergence speed in the later stage of the basic PSO algorithm, which greatly improves its identification accuracy. Finally, through the comparison and analysis of the simulation results, compared with those of the least square (LS) and unmodified PSO identification algorithms, a great improvement in the identification accuracy of the algorithm is achieved. The control effect in the actual control system is also much better than those of the LS and PSO algorithms.

الوصول الحر: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::91e006ef55760750d62577b2002af838Test
https://doi.org/10.1049/iet-csr.2020.0023Test

المؤلفون: Jin Lei Wang, Yafeng Li, Aimin An

المصدر: IET Cyber-systems and Robotics (2020)

مصطلحات موضوعية: parameter estimation, medical robotics, patient rehabilitation, particle swarm optimisation, upper-limb rehabilitation robot system, variable parameter particle swarm optimisation, uncertain parameters, dynamic modelling, upper-limb rehabilitation robots, dynamic parameter identification method, variable parameters particle swarm optimisation, dynamic model, algorithm changes, inertia parameter, learning law, basic pso algorithm, fixed-parameter, identification accuracy, Cybernetics, Q300-390, Electronic computers. Computer science, QA75.5-76.95

الوصف: To solve the problem of uncertain parameters in dynamic modelling of upper-limb rehabilitation robots, a dynamic parameter identification method based on variable parameters particle swarm optimisation (PSO) is developed. Based on the dynamic model of the system, the algorithm changes the inertia parameter and learning law of the basic PSO algorithm from the fixed-parameter to the function that changes with the number of iterations. It solves the problems of small search space in the early stage and slow convergence speed in the later stage of the basic PSO algorithm, which greatly improves its identification accuracy. Finally, through the comparison and analysis of the simulation results, compared with those of the least square (LS) and unmodified PSO identification algorithms, a great improvement in the identification accuracy of the algorithm is achieved. The control effect in the actual control system is also much better than those of the LS and PSO algorithms.

وصف الملف: electronic resource

العلاقة: https://digital-library.theiet.org/content/journals/10.1049/iet-csr.2020.0023Test; https://doaj.org/toc/2631-6315Test

الوصول الحر: https://doaj.org/article/5ea2ca1cb60544e68fd1e63bfee464f3Test