A New Kind of Accurate Calibration Method for Robotic Kinematic Parameters Based on the Extended Kalman and Particle Filter Algorithm

Precise positioning of a robot plays an very important role in advanced industrial applications, and this paper presents a new kinematic calibration method based on the extended Kalman filter (EKF) and particle filter (PF) algorithm that can significantly improves the positioning accuracy of the rob...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on industrial electronics (1982) 2018-04, Vol.65 (4), p.3337-3345
Main Authors: Jiang, Zhihong, Zhou, Weigang, Li, Hui, Mo, Yang, Ni, Wencheng, Huang, Qiang
Format: Article
Language:English
Subjects:
Accuracy
Algorithms
Calibration
Extended Kalman filter
Extended Kalman filter (EKF)
Industrial applications
Kinematics
Mathematical model
Noise
non-Gauss noise
nonlinear system
Nonlinear systems
Parameter identification
particle filter (PF)
Robot kinematics
robotic kinematic parameters calibration
Robots
Service robots
Truncation errors
Viability
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Precise positioning of a robot plays an very important role in advanced industrial applications, and this paper presents a new kinematic calibration method based on the extended Kalman filter (EKF) and particle filter (PF) algorithm that can significantly improves the positioning accuracy of the robot. Kinematic and its error models of a robot are established, and its kinematic parameters are identified by using the EKF algorithm first. But the EKF algorithm has a kind of linear truncation error and it is useful for the Gauss noise system in general, so its identified accuracy will be affected for the highly nonlinear robot kinematic system with a non-Gauss noise system. The PF algorithm can solve this with non-Gauss noise and a high nonlinear problem well, but its calibration accuracy and efficiency are affected by the prior distribution of the initial values. Therefore, this paper proposes to use the calibration value of the EKF algorithm as the prior value of the PF algorithm, and then, the PF algorithm is used further to calibrate the kinematic parameters of the robot. Enough experiments have been carried out, and the experimental results validated the viability of the proposed method with the robot positioning accuracy improved significantly.
ISSN:0278-0046
1557-9948
DOI:10.1109/TIE.2017.2748058