A new calibration method for enhancing robot position accuracy by combining a robot model–based identification approach and an artificial neural network–based error compensation technique

Abstract

Robot position accuracy plays a very important role in advanced industrial applications. This article proposes a new method for enhancing robot position accuracy. In order to increase robot accuracy, the proposed method models and identifies determinable error sources, for instance, geometric errors and joint deflection errors. Because non-geometric error sources such as link compliance, gear backlash, and others are difficult to model correctly and completely, an artificial neural network is used for compensating for the robot position errors, which are caused by these non-geometric error sources. The proposed method is used for experimental calibration of an industrial Hyundai HH800 robot designed for carrying heavy loads. The robot position accuracy after calibration demonstrates the effectiveness and correctness of the method.

A Robotic Calibration Method Using a Model-Based Identification Technique and an Invasive Weed Optimization Neural Network Compensator

“…The HH800 robot [36] model is given in Table 1 and Figure 1. The homogeneous transformation matrix from the robot base frame to the end effector frame can be computed by: The HH800 robot [36] model is given in Table 1 and Figure 1. The homogeneous transformation atrix from the robot base frame to the end effector frame can be computed by:…”

“…The homogenous transformation matrix from the robot 6th frame to the end effector frame: The passive joint position 3 is formed from the joints 2 and 3 as follows [36,47]: The homogenous transformation matrix from the robot 6th frame to the end effector frame:…”

“…The conventional back-propagation neural network (BPNN) [34] is widely adopted by researchers [35] for compensating the unmodeled errors to increasing the precision of the robot. In robotic calibration processing, BPNN is usually employed to construct the relationship between the end effector position and the corresponding joint angle configuration [36][37][38][39]. However, the conventional BPNN has some drawbacks such as getting stuck in local minima and slowing convergence [40].…”

Improvement of Heavy Load Robot Positioning Accuracy by Combining a Model-Based Identification for Geometric Parameters and an Optimized Neural Network for the Compensation of Nongeometric Errors

“…Recently, for nonparametric calibration, some intelligent algorithms, such as genetic algorithm [7,19,20], maximum likelihood estimation [21,22], neural network [23][24][25][26][27][28][29], and various hybrid algorithms [30], have emerged. Among them, neural network has been widely employed to build the relationship, especially nonlinear relationship between inputs and outputs.…”

“…In [26], a multilayer perceptron neural network is utilized to describe the relationship between the joint angles and the corresponding joint errors. Nguyen proposed a technique for the calibration of industrial robots by combining the geometric model-based calibration method and the ANN to identify the kinematic errors, joint compliance errors, and the nongeometric errors [29]. But, at the same time, the BP neural network has the problem that the performance relies too much on the input data and initial values of weights and biases and easily falls into the local optimum.…”