Aircraft manufacturing is a multi-billion dollar worldwide industry ?but one that’s still in search of a completely automated set of solutions. The first ones to achieve that target will surely steal a segment of the industry –with very profitable results.
Researchers at the Fraunhofer Project Group Joining and Assembly FFM (part of the Fraunhofer Institute for Manufacturing Technology and Advanced Materials in Bremen, Germany) are working on a robotic assembly-line system for airplane assembly based on tweaked automotive robots.
If the team is successful, completely automated airplane manufacturing –from CAD design to assembly– may be a reality within a few years, greatly reducing the costly and time-consuming processes currently used in airplane manufacturing.
The Fraunhofer system could see fuselage segments, tail fin and wings on rolling assembly lines being carried past one-armed robots. These robots will then work in succession to bond, drill and mill the parts as they pass, as similar robots do in automotive production processes.
If successful, the project will result in a totally new airplane assembly philosophy, in which entire aircraft are machined ?and their parts bonded together? by teams of small, flexible industrial robots.
Aircraft manufacture is a challenging process. For example, to construct a fuselage, huge parts have to be molded and then tested and measured before components are added using adhesive bonding technologies. Precision machining is then required to complete the part ?and only then can assembly begin.
Despite the enormous size and weight of aircraft parts –individual fuselage segments alone can measure ten meters or more and weigh several tons– the maximum deviation that aircraft manufacturers can tolerate is 0.2 millimeters. Because of the level of accuracy required, assembly is traditionally based on manual operations.
To add to the challenge, aluminum parts are increasingly being replaced by carbon-fiber-reinforced plastic (CFRP) ?an unyielding material that often needs to be assembled under tension by human technicians familiar with the tension levels each part can tolerate.
In order to position these parts accurately, manufacturers use huge production facilities known as ‘assembly cells’ –huge gantries that move along container cranes on steel rails. Assembly cells require massive concrete foundations, are costly and time-consuming to build, and they need to built from scratch for each new type of aircraft under construction.
?What we want to do is to use standard robots like you normally use in the automative industry and enable these standard robots to be highly accurate in order to fit the needs of CFRP assembly processes,? Fraunhofer project manager Gregor Grassl told Robotics Business Review.
That means tweaking automotive robots with improved calibration software and optical laser-based measurement tools. While it is possible to find robots capable of the levels of calibration required for accurate assembly production is, these robots are prohibitively expensive, says Grassl.
Flexibility is key to the proposed design with the tools and software developed being capable of deployment on a wide range of existing automotive robots.
?That’s why we choose to use standard robotics platforms: it could be ABP or Mitsubishi. It could be any robot actually. We want to be highly-flexible in terms of the robot platform. That’s why we are developing software routines, which, in principle, can be adapted to any robot type,? explains Grassl.
It’s easy to teach robots how to work with the ?highly-accurate? parts used in automobile assembly processes, but airplane assembly is a little different says Grassl.
?In automotive assembly you just tell the robot ‘This is the point you should start and this is the point where you end,’ and every metal part which is welded for example will be on the same spot, plus or minus some microns. So, there is no problem with teaching. But with CRFP, every part is unique in terms of dimensions and the position as well. Because it’s a heavy part, it’s flexible or elastic so you cannot just ‘put it on a table’ or it will deform. You have to hold it,? says Grassl.
To combat this problem, Fraunhofer’s proposed robotic assembly line will integrate measurement and calibration tools with flexible handling processes and adaptive tooling techniques.
Currently airplane components are machined by large, inflexible portal milling machines, with measurement and machining taking place at different stations. Fraunhofer’s system is designed to enable small, flexible, adaptive machining with parallel working modules, controlled by measurement systems that can determine real part-position.
Grassl’s team has already developed a flexible component gripper that can deal with the complex geometry of airplane parts. The gripper has a configurable array of suction pads sitting on robust joints with suction pads mounted on a framework structure made of CFRP to provides stability.
The CFRP gripper weighs considerably less than a metal gripper. Because of its low mass, researchers hope that industrial robots will be able to position the gripper and the component with exceptional precision.
And instead of using human technicians to compress or slightly bend the airplane parts to ensure they comply with the aviation industry’s 0.2 mm deviation requirements, the Fraunhofer system will use its gripper and robots to perform this sort of task.
Expect small robots working in assembly teams rather than huge single-function robotics to result from this project, says Grassl.
?We don’t want to have big robots or big machines. That’s why we decided to have automative robots, because they are more flexible and smaller. So, for example, you’re even able to work in parallel on both sides of one part,? says Grassl.
This and other processes will be tested in Fraunhofer’s enormous (80m X 50m X 20m) shop floor in CFK Nord in Stade which has enough space for at least two full-barrel fuselages.
The team will begin construction of the first robotic assembly cell at the end of the year, with a fully-functioning prototype expected to be in operation by the end of 2012.
Fraunhofer aren’t the only people working on automated aircraft assembly, although most initiatives have been somewhat piecemeal due to a focus on single elements of the assembly process rather than complete automation.
The University of Montreal campus, for example, hosts the Aerospace Manufacturing Technology Centre (AMTC), a facility funded by the National Research Council of Canada and Canada Economic Development for Québec regions.
The AMTC focuses on how automation, robotics, and intelligent manufacturing systems –including intelligent surface treatment and automated surface finishing technologies– can bring cost-savings to Canadian aerospace companies. Experts at the facility are researching on low-cost, reconfigurable robotized cells for aircraft component assembly and large-scale machining operations.
The AMTC facility has a large (20? x 20? x 20?) gantry system equipped with multi-axial industrial robots. These include a rail-mounted KUKA 500 robot, a KUKA 210 robot, an ABB 1440 robot, and a PushCorp active force end effector.
In one project, AMTC experts are working with Bombardier Aerospace to create a vision system for drilling sequence and panel inspection, and on the design of robotized cell auxiliary hardware components. A hardware demonstrator to position panels for riveting operations is being developed for another project and orbital riveting of aircraft structures is also being investigated.
In partnership with L3 Com, an automated self-calibration system for industrial robots is being studied. AMTC is also performing a hardware demonstration of automated spar assembly drilling that results in minimum burrs in a collaborative project with Avcorp.
Meanwhile, Spanish-built Roptalmu is a lightweight and portable robot designed by the Industrial Systems Unit at Fatronik-Tecnalia for the Airbus aeronautics company. First coming to prominence in late 2008, the robot was developed to perforate holes in large-scale aeronautic components, such as aircraft wing spars, during their assembly stage.
Roptalmu provides mobility within the assembly process, enabling the aircraft component that’s being worked on to be fixed to its tool holder while the robot moves over the part. The robot’s portability means that the transfer or removal of large-scale parts and tool holders within the manufacturing plant can be avoided, a great advantage compared to traditional assembly set-ups, which require heavy machinery fixed to the floor.
Aircraft manufacturing is a multi-billion dollar worldwide industry ?but one that’s still in search of a completely automated set of solutions. The first ones to achieve that target will surely steal a segment of the industry –with very profitable results.Read More