ABB aerospace seminar

Industrial Robot

ISSN: 0143-991x

Article publication date: 6 March 2009

85

Citation

(2009), "ABB aerospace seminar", Industrial Robot, Vol. 36 No. 2. https://doi.org/10.1108/ir.2009.04936bab.002

Publisher

:

Emerald Group Publishing Limited

Copyright © 2009, Emerald Group Publishing Limited


ABB aerospace seminar

Article Type: News From: Industrial Robot: An International Journal, Volume 36, Issue 2

ABB Robotics held its first open day for aerospace at the UK headquarters in Milton Keynes on September 23, 2008. It attracted delegates from several aerospace companies including Rolls Royce, Airbus UK and G.E Avionics, and also integrators such as RTS Flexible Systems Ltd Robot sales to the car industry have levelled off, and ABB is seeking to address emerging areas such as food and aerospace.

Aircraft manufacturers produce small quantities of highly technical and expensive items, and rely to a large extent on skilled hand-tool operators. However, these people are reaching retirement age, with few young people to replace them. Yann Prigent from ABB France, who specialises in aerospace manufacturing, presented a series of applications showing the robot’s advantages in that sector. These include consistency of product quality, predictability of tool wear and turnaround time, and reduction of material waste. New types of robot application are emerging related to the increasing use of composite materials. ABB has calculated that robots have an operating cost of £5/h compared with £12-15 for a human worker when overheads are taken into account. Robots need no employer tax payments, and reduce the cost of heating, lighting and health and safety management. Robots can undertake health-threatening tasks safely, and are not susceptible to repetitive strain injury or vibration white finger.

In France, ABB’s customers are aeroplane manufacturers such as Dassault Aviation or sub-suppliers like the Safran Group, which produces engines for airbus. Robots are used in repetitive applications in both surface treatment and drilling. In the surface treatments of shot peening, surface coating and painting, robots exhibit superior path repeatability compared to manual workers. They tend CNC machines in precision grinding and milling of turbine blades, and are used in assembly for drilling and riveting. Coriolis Composites is using an IRB 6650/2 articulated robot mounted on a 20 m travel track to place fibre material in moulds. This has led to four patents in guiding, tensioning and fibre placement techniques. ABB is also involved in a development project for ultra-sonic welding of composite structures, an operation that requires high forces. Prigent stressed the robot’s adaptability and its competence in careful handling of expensive parts.

Force Control

Where there is a skills shortage, robot programming can be challenging, and ABB demonstrated the ability of its Force Control Technology to simplify the robot teach process (Figure 2). A 6-axis sensor load cell, visible at the top left of Figure 2 and unique to ABB, continuously monitors the force exerted by the end effector. Close feedback to the arm motors ensures that it always applies the correct force, for instance in the assembly of delicate parts. By setting the required force along X- and Y-real-world axes, the robot can be made to move along the perimeter of a workpiece, pressing against it all the time. This is useful not only in operations such as deburring, but also in the process of teaching the robot. From just a few manually taught locations around the perimeter of a workpiece, the robot can learn the detailed path for itself by following the perimeter with constant contact force. With this technique, it takes only 1 or 2 h to program the shape of a turbine blade, compared with 1 week by standard means. In the same way, the robot can learn the shape of a complicated 3D object and perform polishing operations with great consistency.

 Figure 2 Using Force Control, the operator can teach the robot a few
locations around the perimeter of a workpiece

Figure 2 Using Force Control, the operator can teach the robot a few locations around the perimeter of a workpiece

Force Control enables the robot to handle parts by “feeling its way” rather than hard-programmed location control. It can gently move tools into position whilst saving the cost of fixtures and the programming time needed for more precise positioning. It handles and positions the toothed rings and hubs in clutch assembly, and inserts shafts into semi-precision holes by tracing a spiral path across the surface until the shaft slots in.

Force Control can be used in two different modes. Using FC Pressure software, the robot endeavours to apply constant pressure to a surface, changing its position as necessary. This is used in deburring, polishing, grinding and cleaning. FC SpeedChange, on the other hand, makes the robot follow a precise path, operating at maximum process speed where possible but slowing down when the machining forces become too high. This is helpful for milling along the edge of a workpiece, and grinding unevenly distributed material on cast surfaces. Force Control allows robots to tackle jobs where they would otherwise break tools, damage the workpiece or overload and stop.

Vision guidance

Mike Badger of ABB worked hard to convince his audience that robotic vision guidance is different from machine vision. Perhaps, he was trying to reduce its perceived complexity, but I found the distinction subtle and unnecessary. He claimed that TrueView vision guidance makes 3D measurements from a single image, by taking account of the size of the target as well as its position. (Of course, this is only possible by calibrating the depth dimension, and that this is done by capturing reference images!) Vision guidance, however it is defined, allows the robot to adapt to the manufacturing environment (Figure 3), accepting randomly oriented or disordered parts and making enormous savings on alignment and positioning fixtures. It transforms the capabilities of the robot, coping with variations in part location and type, and detecting detects. Using geometric pattern recognition, the vision system locates features within the target and corrects the position and orientation of the robot end effector to deal with randomly presented workpieces. One application is automated engine assembly. Toyota, Ford, Honda GM, and Crysler all use TrueView. But, what of the aerospace industry? How is it used there? According to Yann Prigent, vision is currently used for part identification and inspection, and for drilling and riveting.

 Figure 3 TrueView equips the robot with vision guidance

Figure 3 TrueView equips the robot with vision guidance

TrueView uses eVisionFactory software incorporating algorithms relevant to and optimised for robot use. System calibration and object identification are automated, making the system easy and quick to set up. Various lighting packages are available to make feature-finding simple and reliable: for example, infrared illumination and filtering are useful in variable ambient light conditions. Fast communication of data-intensive images to the robot controller is essential, and TrueView closely integrates into IRC 5, S4C+ and S4C controllers for use with all current ABB robots.

RobotStudio and Remote Service

At Milton Keynes, a rolling programme of training courses operates for 48 weeks per year, and is attended by university students as well as industrial personnel. ABB’s simulation and off-line programming software RobotStudio (Figure 4) is found in many university engineering laboratories. The package is supplied free of charge to bona fide educational institutions, and a basic version will soon be available for free download. RobotStudio allows programs to be modified without production downtime. In preparing new applications, it checks accessibility and calculates cycle times, establishing whether the procedure is viable, and verifying tooling and fixtures at the design stage. The preparatory work is carried out on a PC, and downloaded to the controller when it is ready to go on line. Ian Schofield presented a case study involving plasma spraying. The part was geometrically so complex that it could not be programmed by normal teach methods. But by importing the CAD data into RobotStudio, a very accurate program was quickly generated, and a better position was found for the robot relative to the part.

 Figure 4 ABB’s RobotStudio off-line programming software saves downtime
in industrial robots and is widely used in universities to train robot engineers

Figure 4 ABB’s RobotStudio off-line programming software saves downtime in industrial robots and is widely used in universities to train robot engineers

Technical support and engineer call-out are important in maintaining production efficiency. ABB launched its Remote Service system in 2007, taking a proactive maintenance approach (Figure 5). By monitoring customer’s robots during operation, ABB can predict when problems are likely to occur, and can dispatch a service engineer as necessary. A GPRS transmitter can alert factory personnel via a text message. The system advises the customer when parts need replacing. Alan Lashbrook told me that the system costs about £1,125 per robot per year, including full telephone support and access to ABB parts store. It quickly pays for itself through reduced downtime.

Figure 5 ABB’s Remote Service system monitors the condition of customers’ robots during operation for preventative maintenance and field service response

Related articles