Software simulation platform for learning techniques for identifying kinematic parameters of industrial robots

Author(s):  
Jorge Santolaria ◽  
Juan Aguilar ◽  
David Samper ◽  
Francisco Brosed ◽  
Ana Majarena
2020 ◽  
Vol 900 ◽  
pp. 35-43
Author(s):  
Yunn Lin Hwang ◽  
Jung Kuang Cheng ◽  
Thanh Dat Pham

The simulation and application of industrial robots has developed very quickly in recent decades. Along with the development of computer science, a lot of softwares to perform dynamic simulation have been created. The results of simulation can be used for layout evaluation, kinematic, dynamic study, off-line programming to avoid obstacle and for design mechanical structure of robots. A co-simulation of 2R industrial robots have been performed by Recurdyn and Matlab. The input parameters are executed under Matlab, and then exported to Recurdyn environment. Kinematic parameters will be executed by RecurDyn then exported to Matlab. The main tasks of this paper are performing 2R robotic manipulator kinematic simulation in two postures with the same trajectory and the same time. Thus, the result of simulation can be compared with theories. Finally, a real 2R robot model was used to verify the trajectory with CAE simulation.


2014 ◽  
Vol 889-890 ◽  
pp. 1136-1143
Author(s):  
Yong Gui Zhang ◽  
Chen Rong Liu ◽  
Peng Liu

For an industrial robots with unknown parameters, on the basis of preliminary measurement and data of the Cartesian and joints coordinates which are shown on the FlexPendant, the kinematic parameters is identified by using genetic algorithms and accurate kinematics modeling of the robot is established. Experimental data could prove the validity of this method.


2007 ◽  
Author(s):  
Meng Li ◽  
Meng Yang ◽  
Dong Liu ◽  
Peng Li

Author(s):  
Firas Zoghlami ◽  
Philip Kurrek ◽  
Mark Jocas ◽  
Giovanni Masala ◽  
Vahid Salehi

Abstract The use of flexible and autonomous robotic systems is a possible solution for automation in dynamic and unstructured industrial environments. Pick and place robotic applications are becoming common for the automation of manipulation tasks in an industrial context. This context requires the robot to be aware of its surroundings throughout the whole manipulation task, even after accomplishing the gripping action. This work introduces the deep post gripping perception framework, which includes post gripping perception abilities realized with the help of deep learning techniques, mainly unsupervised learning methods. These abilities help robots to execute a stable and precise placing of the gripped items while respecting the process quality requirements. The framework development is described based on the results of a literature review on post gripping perception functions and frameworks. This results in a modular design using three building components to realize planning, monitoring and verifying modules. Experimental evaluation of the framework shows its advantages in terms of process quality and stability in pick and place applications.


2019 ◽  
Vol 141 (11) ◽  
Author(s):  
Gino Angelini ◽  
Alessandro Corsini ◽  
Giovanni Delibra ◽  
Lorenzo Tieghi

Abstract The main intent of this work is the exploration of the rotor-only fan design space to identify the correlations between fan performance and enriched geometric and kinematic parameters. In particular, the aim is to derive a multidimensional “Balje chart,” where the main geometric and operational parameters are taken into account in addition to the specific speed and diameter, to guide a fan designer toward the correct choice of parameters such as hub solidity, blade number, hub-to-tip ratio (HR). This multidimensional chart was built using performance data derived from a quasi-3D in-house software for axisymmetric blade analysis and then explored by means of machine learning techniques suitable for big data analysis. Principal component analysis (PCA) and projection to latent structure (PLS) allowed finding optimal values of the main geometric parameters required by each specific speed/specific diameter pair.


2019 ◽  
Vol 16 (5) ◽  
pp. 172988141988307 ◽  
Author(s):  
Yahui Gan ◽  
Jinjun Duan ◽  
Xianzhong Dai

Calibration of robot kinematic parameters can effectively improve the absolute positioning accuracy of the end-effector for industrial robots. This article proposes a calibration method for robot kinematic parameters based on the drawstring displacement sensor. Firstly, the kinematic error model for articulated robot is established. Based on such a model, the position measurement system consisting of four drawstring displacement sensors is used to measure the actual position of the robot end-effector. Then, the deviation of the kinematic parameters of the robot is identified by the least-squares method according to robot end-effector deviations. The Cartesian space compensation method is adopted to improve the absolute positioning accuracy of the robot end-effecter. By experiments on the EFORT ER3A robot, the absolute positioning accuracy of the robot is significantly improved after calibration, which shows the effectiveness of the proposed method.


Sign in / Sign up

Export Citation Format

Share Document