QMT Features: October 2017
Calibration raises robot performance
As robots move into more demanding applications calibration becomes more and more of an issue for accuracy says API

Robots have become common place in the manufacturing environment. They are commonly used for simple tasks which are repetitive, and do not require high accuracies. Some examples include pick and place, arc-welding, and part inspection. The major factors in the popularity of robots are their low price, small footprint, and versatility. While these features are critical for the robotics market, they introduce challenges for accuracy and metrology.

Consider a CMM. There is low thermal growth, each axis is independent, and the gantry design which is typically employed offers great stability. The classic robot design offers none of these considerations. Each joint has high thermal variation, and the tool center point is at the end of all six joints, creating considerable variation and sag at the TCP. Additionally, the small footprint of the robot provides for a very poor foundation for the system.

Despite this, robots are known for having high repeatability. This is achieved through the robots’ manufacturing, advanced controllers, and high accuracy encoders. Taking advantage of the repeatability, robot performance can be improved.

Robot Performance
Robot performance can be described, generally, in two parts; positioning and path travel. Robot positioning depends on the controller parameters, such as the home position of the robot, its payload settings, tool offsets, amongst other parameters. Most of these inputs are set by the user, but there are some which are set by the OEM, such as the joint lengths, encoder positions, gear ratios, and other constants the robot uses to calculate its positioning.  The robot path is largely determined by how it is programed, and where the start- and end-point of the travel is. In order to quantify robot performance scientifically, ISO9283 was introduced. After a few iterations, the 1998 edition was written, and has remained the standard with only minor addendums. This standard fully defines robot performance with 14 testing procedures.

Measuring a robot in compliance with the ISO standard has some obvious benefits such as knowing your robot is performing as expected against an internationally recognised and approved standard. Additionally, quantifying the robot’s capabilities helps customers looking to purchase a robot match a robot with their intended use. One advantage which should be noted is that the ISO defines testing procedures. To understand this, consider running a program from the controller to test path accuracy. The program needs to be written in such a way that the robot is told what path to take. If there are only points in the program, this will not suffice to test path accuracy. Similarly, one needs to account separately for the orientation of the tool when considering the path. The ISO standard not only defines what measurements to make, but defines the procedures for testing.

Robot Calibration

After the robot performance is defined, it is natural to ask how to make the robot performance better. The industry has been moving towards a solution known as Denavit-Hartenberg, or DH, Calibration. The DH model of a robot is made up of a four by six matrix of values, two rows of distances and two of angles. In general, Theta is the joint angle between the joint and its preceding one, D is the joint offset between the joint and the preceding one, A is the link length, and Alpha is how much the joint is twisted compared to the preceding joint.

Robot Metrology Solutions

API has long been known for its laser trackers, and innovative measuring solutions. API has now added Robot Metrology Solutions (RMS) to its line of measuring techniques. RMS is a software designed to integrate API Laser Trackers into the robot metrology system. It is comprised of the Robot Performance Measurement (RPM) and DH Calibration modules.
The RPM module measures robots in accordance with ISO9283:1998. The software will first walk the user through setting up the work volume as prescribed by the ISO. This procedure ensures the optimal poses will be used for measuring the robot performance. Additionally, the software will report these poses in coordinate format, allowing the user to enter the points directly into the controller memory for eases and consistency. After the setup, a project file is created.

All of the 14 measurement procedures are available to measure from the project. RPM generates the measurement positions, making programing the test simple. The software already has the nominal positions saved in the project tile, so no further setup is required. As the robot moves to these positions, the tracker will measure three SMRs on the robot tool tip, collecting 6 degree of freedom data about the robot performance. The measurements are easy to run, and the software will report the results as specified by the ISO.

The DH Calibration Module is also easy to use. The robot is first programed to move to any number of locations. There is a minimum of 30 positions, but the more positions used, the better the calibration will be. As the robot moves these positions, the tracker will measure the robot. After the measurement process is complete, the tracker data and robot data are loaded into the software. The software will run inverse kinematics, and generate updated DH parameter which can be utilized by the user.


Basic robot performance is defined by robot metrology. By using a software system like API’s RMS, the process can be straightforward, quick, and informative. By using quantified performance results, a decision can be made to calibrate the robot. API’s RMS software also offers a calibration process which can provide the end user with updated DH parameters which will improve robot performance.
You can now view all QMT Magazine issues on your favourite tablet or smart phone.
Download the free Quality Manufacturing Today App from the Apple iTunes App Store or from QMT Magazine on Google Play.

Rob Tremain Photographer
Click above to see full page display and links to QMT articles.
View Olympus video>>
Control logo
Bowers logo
Mitutoyo logo