Experiment name: ENDORSE

Manufacturing end-user: Enikon Aerospace d.o.o.

The Business Sector

The aircraft manufacturing industry is bracing itself for an anticipated order boom in the next 20 years. Industry experts observe that air traffic doubles every 15 years. As a result, experts predict that by 2035 there will be an estimated demand for 33,000 – 40,000 new passenger and freighter aircrafts at a value of USD 5.2 -5.9 trillion. Along with capacity expansion in the airline industry, there will also be a need to retire and replace older aircraft models. ENIKON Aerospace d.o.o. (EnAe) is the leading company in the world within the niche market of outsourcing of preparation and final painting of molded interior parts for commercial and business aircraft.

With over 10 30 years of experience ENIKON Aerospace d.o.o has special expertise in finishing and painting composite parts and was the first company in the industry to shift from solvent to water-based paints, and still remains the technological leader in this area. In addition, Enikon has experimented with diversification into other sectors where high quality is a requirement, and potential to achieve high margins, such as the finishing and painting of medical devices.Within the aerospace industry, Enikon provides surface preparation, painting and finishing of interior commercial aircraft parts such as sidewall panels, ceiling panels, passenger/service/utility panels and various other interior components for commercial aircraft. Enikon typically supplies Tier One aerospace suppliers, who in turn supply Original Equipment Manufacturers (OEM’s), such as Airbus, Boeing, Bombardier, Sukhoi and Embraer. Enikon is now making inroads into automation. For example, it has created robots to perform certain grinding and finishing functions, allowing it to produce even more consistent quality with a fraction of the labor cost. In ENDORSE, EnAe will be involved in consulting partners during software development, as well as testing and system integration. ENDORSE_main2

The Challenge

In order to respond to an ongoing skilled labor shortages and demand for high quality products, EnAe has begun with the automation of grinding process, which is vital for surface preparation prior to the high-quality painting finish. Automation of this process has several benefits: i) consistent quality of the final product, ii) no human errors, iii) the wastage of material is reduced to minimum, and iv) the exposure of employees to dusty atmosphere is minimized.

Although the current robotic grinding work cell can process up to 75% of part’s surface, the time the part spends being processed by the robot is only 8% of the total processing time. Other 92% of the time, the part is handled and processed by human workers. Since the grinding produces a very fine dust, the most important goal of this project is to shorten the time a human worker needs to spend in such hazardous conditions. To that end, an additional robot will be introduced into the process. Use of a more agile and human friendly robotic manipulator in the process will result in the increase of the percentage of a parts area that is processed by robots, as the human workers will be able to teach the robot how to grind more complex parts. In order to shorten the time a human worker processes a part, automated quality inspection system using the aforementioned agile robot equipped with lasers will be developed.

ENDORSE_main

The solution:

ENDORSE experiment builds upon the HORSE framework in order to deploy a semi collaborative industrial manipulator in the process of sanding composite parts for aviation industry. The main challenges of the project were to build a framework that will enable a standard industrial manipulator to operate in a collaborative environment. The four cornerstones of our framework include:

The compliant control algorithm, which implements an impedance-based robot control algorithm to improve the quality of surface treatment. Being able to offer compliant 6D motion of the whole robotic manipulator allowed the end-user to increase the amount of surface of the products and the types of products that can be treated with a robotic manipulator;

Automated quality inspection system scans the products surface using a linear scanner and calculates the roughness of the surface, localizing the parts of the product that have not been treated properly. It provided the end user with the means to analyse and quantify the quality of each product exiting the production line.

Human oriented machine interface built on top of the compliant control algorithm enables the workers in the factory, accustomed to manual labour, to show the robot how to treat specific parts of the surface. This intuitive concept avoids tedious offline path planning, where it is often hard to capture the optimal motion to treat the most complicated surfaces.

Human safety system utilizes a standard LIDAR sensor to detect objects around the robot. Our system relies on the concept of zone detection. When an actor enters a specific zone around the robot workspace an appropriate action is selected for the robot, thus keeping humans from harm.

Once each hardware and software component were developed, we proceeded with system integration. Working closely with the end user, operators and inspectors allowed us to benchmark the results of the ENDORSE system against the current production line. This clearly showed that we can increase the amount of surface treated by the robot and variety of products, as well as precisely locate treatment imperfections.

The integrated system consists of three major components: HORSE based supervisor station, ENDORSE grinding system based on Kuka KR robot and a Graphical user interface for offline trajectory planning. All the components are interconnected through technologies provided within ROS and HORSE framework. Using the HORSE core technologies, we implemented a Business Process Model for the end user and connected it to the proposed framework. Ultimately, this enables the end user to track the production line in real time.

Partners:

ENDORSE_icent ENDORSE_larics
ENDORSE_fer ENDORSE_enikon