项目来源

俄罗斯科学基金(RSF)

项目主持人

Maloletov Alexander

项目受资助机构

Autonomous noncommercial organization of higher education"Innopolis University"

立项年度

2022

立项时间

未公开

项目编号

22-41-02006

项目级别

国家级

研究期限

未知 / 未知

受资助金额

未知

学科

MATHEMATICS,INFORMATICS,AND SYSTEM SCIENCES

学科代码

01

基金类别

未公开

关键词

Робототехника ; Искусственный Интеллект ; Роботизированная формовка ; Метод конечных элементов ; Robotics ; Artificial Intelligence ; metal forming ; robo-forming ; Finite Element Analysis

参与者

未公开

参与机构

未公开

项目标书摘要：nnotation:Incremental sheet metal forming(ISF)is a dieless metalworking process that relies on a progression of localized deformation.The desired shape is obtained by applying localized pressure with a simple rigid tool that moves across a metal sheet.This technique offers the possibility to fabricate complex 3D shapes which were previously difficult or sometimes not possible to produce through conventional processes like metal spinning.It is especially cost-effective in low batch size production since costly dies are not required.It is one of the most promising technologies for aerospace,automotive,and medicine.However,conventional incremental metal forming centers have limits on available workspace dimensions and thus difficult for big workpiece production.Therefore robot-based incremental sheet metal forming or robo-forming is recently being investigated for the increased flexibility in tool motion and size of the workspace.However,in such a system,a large number of variable factors can affect the quality of the end product.The accuracy of the final product is determined by the two aspects,i.e.,the accuracy of the robots that are involved,and the effect of material undergoing deformation.Though there are many contributing factors,the most important aspect influencing the quality of the final product are robot compliance under external loading and aspects of sheet metal forming like spring back and strain to harden.These issues have coupled effect on the quality of the final product and will be investigated together in the scope of the project.The overall goal of the project is to integrate this knowledge on the level of robot trajectory planning while framing a task for robot-based metal forming.To achieve the desired goals traditional model-based approaches will be enhanced with benefits from artificial intelligence(AI)based methodologies for metal behavior model design in the process of incremental sheet metal forming,robot stiffness model design for compliance error compensation,and advanced optimized path planning for the development of components using robot forming.
        Modeling of incremental sheet forming is complex and solution time for complete simulations of a forming operation is currently many times greater than the time required to form the product.Therefore,high accuracy forming will only be achieved either by repeated trials of tool paths with corrections based on measuring the errors in the formed parts,or by use of some form of online shape measurement and feedback control via artificial intelligence(AI)algorithms,to modify the tool path in realtime.AI focuses on the use of data and algorithms to imitate the way that humans learn,gradually improving the accuracy of the process.An AI system is trained with many examples relevant to the task,and it finds statistical structure in these examples that eventually allows the system to come up with rules for automating the task.The major drawback in the robotic incremental sheet forming process is a large number of influencing factors like geometrical inaccuracy due to the spring back effect,limited stiffness of the robotic setup,robot redundancy,etc.These issues in end-product quality can only be dealt with by in-process monitoring and feedback mechanisms governed by AI algorithms.
        Successful realization of the project requires the involvement of experts from material science and metal forming(IIT Madras,IIT Ropar,IIT Bhilai,India),robot modeling,and control(Innopolis University,Russia),and Artificial Intelligence(Innopolis University,Russia and IIT Madras,India).
        Expected results:Anticipated Results
        The following outcomes are expected from the project:
        1.Methods of reducing the stiffness model complexity,without compromising accuracy,to simplify the stiffness modeling of heavy-duty industrial serial robots used in sheet metal forming.
        2.Algorithms to select the best poses for identification of the reduced model of elasto-static parameters.
        3.Methods of planning the robot trajectory,given the force and spring back effect,from simulation studies.
        4.Artificial Intelligence based algorithms to enable real-time path correction using information from the robot mounted sensors.
        5.Hybrid Artificial intelligence-based model-free and model-based methods for robot calibration.
        6.Strategy for calibration and subsequent coordinated motion of two robots while incorporating redundancy based path planning with an objective to attain required shape of metal sheet,simultaneously avoiding the singularity,increasing rigidity etc.
        These solutions are required to transfer the technology of robotic metal forming from the laboratory to mass industrial application.The realization of the project will create technologies that will be in demand in airspace industry,medicine,shipbuilding and other areas.
        The obtained results will be published in 18 scientific papers indexed in the Scopus database,among which at least four papers in journals included in the Q1.

Application Abstract: Annotation:Incremental sheet metal forming(ISF)is a dieless metalworking process that relies on a progression of localized deformation.The desired shape is obtained by applying localized pressure with a simple rigid tool that moves across a metal sheet.This technique offers the possibility to fabricate complex 3D shapes which were previously difficult or sometimes not possible to produce through conventional processes like metal spinning.It is especially cost-effective in low batch size production since costly dies are not required.It is one of the most promising technologies for aerospace,automotive,and medicine.However,conventional incremental metal forming centers have limits on available workspace dimensions and thus difficult for big workpiece production.Therefore robot-based incremental sheet metal forming or robo-forming is recently being investigated for the increased flexibility in tool motion and size of the workspace.However,in such a system,a large number of variable factors can affect the quality of the end product.The accuracy of the final product is determined by the two aspects,i.e.,the accuracy of the robots that are involved,and the effect of material undergoing deformation.Though there are many contributing factors,the most important aspect influencing the quality of the final product are robot compliance under external loading and aspects of sheet metal forming like spring back and strain to harden.These issues have coupled effect on the quality of the final product and will be investigated together in the scope of the project.The overall goal of the project is to integrate this knowledge on the level of robot trajectory planning while framing a task for robot-based metal forming.To achieve the desired goals traditional model-based approaches will be enhanced with benefits from artificial intelligence(AI)based methodologies for metal behavior model design in the process of incremental sheet metal forming,robot stiffness model design for compliance error compensation,and advanced optimized path planning for the development of components using robot forming.
        Modeling of incremental sheet forming is complex and solution time for complete simulations of a forming operation is currently many times greater than the time required to form the product.Therefore,high accuracy forming will only be achieved either by repeated trials of tool paths with corrections based on measuring the errors in the formed parts,or by use of some form of online shape measurement and feedback control via artificial intelligence(AI)algorithms,to modify the tool path in realtime.AI focuses on the use of data and algorithms to imitate the way that humans learn,gradually improving the accuracy of the process.An AI system is trained with many examples relevant to the task,and it finds statistical structure in these examples that eventually allows the system to come up with rules for automating the task.The major drawback in the robotic incremental sheet forming process is a large number of influencing factors like geometrical inaccuracy due to the spring back effect,limited stiffness of the robotic setup,robot redundancy,etc.These issues in end-product quality can only be dealt with by in-process monitoring and feedback mechanisms governed by AI algorithms.
        Successful realization of the project requires the involvement of experts from material science and metal forming(IIT Madras,IIT Ropar,IIT Bhilai,India),robot modeling,and control(Innopolis University,Russia),and Artificial Intelligence(Innopolis University,Russia and IIT Madras,India).
        Expected results:Anticipated Results
        The following outcomes are expected from the project:
        1.Methods of reducing the stiffness model complexity,without compromising accuracy,to simplify the stiffness modeling of heavy-duty industrial serial robots used in sheet metal forming.
        2.Algorithms to select the best poses for identification of the reduced model of elasto-static parameters.
        3.Methods of planning the robot trajectory,given the force and spring back effect,from simulation studies.
        4.Artificial Intelligence based algorithms to enable real-time path correction using information from the robot mounted sensors.
        5.Hybrid Artificial intelligence-based model-free and model-based methods for robot calibration.
        6.Strategy for calibration and subsequent coordinated motion of two robots while incorporating redundancy based path planning with an objective to attain required shape of metal sheet,simultaneously avoiding the singularity,increasing rigidity etc.
        These solutions are required to transfer the technology of robotic metal forming from the laboratory to mass industrial application.The realization of the project will create technologies that will be in demand in airspace industry,medicine,shipbuilding and other areas.
        The obtained results will be published in 18 scientific papers indexed in the Scopus database,among which at least four papers in journals included in the Q1.