In recent years, major international manufacturers have begun to try to introduce artificial intelligence (AI) technology into intelligent production line applications. For example, Epson demonstrated autonomous artificial intelligence dual-arm robot in 2017; GPU manufacturer deep-rooted artificial intelligence market At the 2018 Taipei International Computer Show (Computex 2018), NVIDIA announced the launch of the AI chip Jetson Xavier for robotics, which is expected to enable industrial robots to work side by side with humans safely and for various changes. Adaptation.
In 2018, one of the most obvious advances in smart manufacturing was the dramatic advancement of sensor technologies such as 3D Vision, radar technology and capacitive technology. At the same time, the market's increased demand for shorter product cycles and flexible manufacturing also prompted robots. Increased flexibility and performance in automated processes.
Seven-axis robotic arm movement space is better than six axes
The development and application of multi-axis robot arm can save manpower, replace human beings into dangerous working environment and improve the quality of precision machining. The precision of the robot arm and the low error characteristics have advantages for the mastery of product quality and can be effective. Reduce the time and labor required for quality control.
Xiong Zhimin (Fig. 1), manager of IKK Machinery and Manufacturing Systems Research Institute of ITRI, pointed out that it is most suitable for importing robots in applications with high manpower requirements, poor working environment and simple application requirements. For example, 3C product manufacturing plants with large assembly requirements High-risk metal foundries are quite suitable for importing robots to replace manpower. However, according to various manufacturers in Taiwan, the import ratio is estimated to have not exceeded 30%, and there is still considerable room for growth.
Figure 1 Xiong Zhimin, manager of the IEK Machinery and Manufacturing Systems Research Institute of the ITRI, pointed out that it is most suitable for importing robots in areas with high manpower requirements, poor working environment and simple application requirements.
In order to achieve a more flexible arm movement space, in addition to continuing to promote the relatively mature six-axis robotic arm, each manufacturer has also begun to develop a seven-axis robotic arm. NEXCOM is able to jointly publish a seven-axis robot with American software company Energid Technologies in 2017. Lin Chongji, deputy general manager of Xinhan Intelligent Technology Industrial IoT Intelligentization Integration Division, pointed out that the translational mechanical arm can only achieve four-axis motion at most. If it is to replace manpower, the articulated robot can do it. Flexibility is unmatched by the translation. The seven-axis robot can achieve a more flexible motion space than the six-axis. In the face of smart manufacturing, the future will move toward a smaller and more flexible manufacturing trend. It seems that the seven-axis robot can reach the changing production process. The flexibility required at the time. However, the seven-axis robot is still a fairly high-end product, and its high unit price is not affordable for all industries. It seems that the automobile manufacturing industry will be the first application field.
Cost-effective / professional talent for the import of the biggest consideration
In fact, it is not just a high-end seven-axis robot. In the process of introducing various types of robotic arms into the manufacturing line, the cost considerations are the biggest pain points of the manufacturers. Lin Chongji believes that the price of all kinds of robotic arms must be reduced at least. 30%, its penetration rate can be further improved.
Xiong Zhimin also believes that for Taiwanese manufacturers, Return On Investment (ROI) is the biggest consideration when purchasing equipment. ROI must combine capacity, electricity consumption, power-on time for machine wear, etc. And many other factors, small and medium-sized manufacturers or need to assist the evaluation by the system operator.
In addition to cost considerations, many manufacturers also pointed out that professionals are very scarce in the field of intelligent manufacturing. In the process of industrial transformation intelligent manufacturing, various types of machinery and equipment are slowly replacing manpower, but this is limited to high risk. Sexual, highly repetitive work. In different production areas, many professional knowledge can not rely on the cooperation system manufacturers, the industry must personally evaluate, in order to determine whether the introduction of the robot arm can meet the manufacturing needs. In the future, artificial intelligence begins to import In the production line, more relevant talents are needed to assist in deep learning training. However, for professionals with artificial intelligence and big data analysis, the industrial field is often not the primary employment choice.
Therefore, Bosch assisted the National University of Success in creating the “Intelligent Manufacturing Innovation Center” (Figure 2). It is a joint venture to develop precision machinery and equipment, robotics, industrial Internet of Things, and big data applications. Key technologies such as cloud computing and automation to achieve the goal of industry-university integration. Alliance schools will also use this training machine to plan a series of problem-oriented practical courses that are rich and complete, so that students can Hands-on verification of Industry 4.0 intelligent manufacturing theory and creative thinking, and then achieve the purpose of eliminating the gap between learning and learning.
Figure 2 Taiwan Bosch assisted the National Cheng Kung University to build the “Intelligent Manufacturing Innovation Center” of the first industrial 4.0 education center in Taiwan.
Lin Chongji also pointed out that in 2018, Xinhan Technology has promoted its own educational robotic arm to more than 20 colleges and universities for teaching use. It also provides teaching materials and support teachers to help students learn to operate robotic arms and talents. After getting out of the education system, he can immediately connect with the industry. Lin Chongji stressed that if there is a lack of relevant talents, the overall industry will be difficult to promote.
Wang Shenglin (Fig. 3), deputy associate of Siemens Digital Factory Division, also believes that although artificial intelligence is one of the hottest topics in the industry, for manufacturing applications, there is still a long period of time to see the proliferation of artificial intelligence. Taking the robot arm as an example, the arm moves safely and correctly. It is the two premise that the production line is introduced into the arm. However, whether artificial intelligence has such a high degree of reliability can make people feel comfortable with the work of arm movement control. It is still a big problem to deal with artificial intelligence. In addition, manufacturers often encounter many problems that require professional system integrators (SI) to assist in the process of importing their arms, because manufacturers often do not have arm control, programming. Relevant knowledge and ability. This problem will become more prominent after entering the AI generation, and the work of SI will become more challenging.
Figure 3 Wang Shenglin, deputy associate of Siemens Digital Factory Division, believes that in terms of manufacturing applications, there is still a long way to go to see the proliferation of artificial intelligence.
For Siemens, artificial intelligence has entered a large number of applications in the medical and energy fields. However, for the manufacturing industry, it is still in the research stage. There is no standardized product solution. However, some have made faster manufacturing. The industry has already begun to have artificial intelligence related requirements. For these leading customers, Siemens is working with the project to jointly introduce artificial intelligence into production online.
Green leaves become red flowers
Many of today's discussions on smart manufacturing often emphasize unmanned factories, light-off factories, and fully automated production lines, emphasizing that the production line is fully controlled by machinery, and all of the staff are exiting. However, in fact, most manufacturing industries are still in the industrial stage. Below 3.0, in the production of most imported robots online, in fact, professional staff is still the core.
In the opinion of Taiwan's Bosch Rexroth factory automation sales associate Chen Junlong, the robotic arm is only an auxiliary tool; especially on flexible conveyor belts, the staff must be the key player in production online.
Chen Junlong further explained that in a small number of diverse production lines, different items may be interspersed. Through RFID sensing, the equipment is allowed to know the items brought in by the conveyor belt. The robot arm must grab different assembly parts according to different items. At the same time, the parameters of the machine will change, and the process will be displayed on the billboard. In this process, the production online staff must take care of the responsibilities and cooperate with the robot. The main job of the robot is to be responsible for repeatability. Actions to reduce the burden on personnel and prevent human error.
Esben H. stergaard (Fig. 4), founder and chief technology officer of Universal Robots, believes that the first step in upgrading smart manufacturing is to purchase collaborative robots to assist in simple application. For example, using computer numerical control machine (CNC) support. Carrying out work, etc., to accumulate experience of automation import for enterprises, and to start from this, continue to improve the automation process.
Figure 4 Esben H. Østergaard, founder and chief technology officer of Universal Robots, believes that the first step in upgrading smart manufacturing is to purchase collaborative robots to assist in simple application.
On the other hand, artificial intelligence technology has recently received a lot of media attention. stergaard pointed out that like a drill or a lathe, a robot is a tool, and there are many advanced software inside the robot that can be called artificial intelligence. In the future, the robot will It can be a predictable and reliable manufacturing tool. Moreover, the robot will also be the naturally generated data hub in the manufacturing plant. They have the networking and computing power to collect data of all relevant applications in the manufacturing environment. Meanwhile, the robot The high applicability in the plant also makes it easy to perform different tasks.
To import a collaborative robot, it is best to start with a simple application, such as a cooperative robot to assist in handling (Machine Tending), an automated guided vehicle (AGV), or various applications for loading and unloading, workpiece displacement (eg Locking screws, workpiece grabbing) are all areas that are easier to import.
However, Wang Shenglin believes that the future robotic arm will certainly not only be used for workpiece grabbing, handling such relatively simple work. When the arm is combined with processing equipment, such as tools, grinding wheels, nozzles, the arm will be able to cut the workpiece directly. Surface treatment such as sanding or painting, which poses a certain degree of threat to existing machine tools.
Compared with the machine tool, the biggest advantage of the robot arm is that it can be operated flexibly. One arm can process the workpiece from various angles, but the machining angle of the machine tool is limited, if the workpiece is to be processed more complicatedly. The program may have to be completed with two machine tools. But in some cases, one arm can do two machine tools.
However, to achieve this advanced application, the biggest challenge is to control system integration. At present, robotic operators usually have their own control systems, while other field devices also have dedicated control systems. How to integrate two different control systems, let The ability of the arm to collaborate with other field devices is one of the main challenges facing the industry. Siemens does not provide a robotic arm, but it can work with the major arm manufacturers to achieve arm and other sites on a single control system. Equipment cooperation, such as Kuka's mxAutomation is the Siemens solution, which can be used to achieve this type of cross-machine integration with a single control system.
In addition to easier integration, Siemens' own control system also has some unique features that can be used to directly process the workpiece with the arm. In traditional workpiece grabbing operations, the arm movement settings are usually linear point-to-point motion. However, if the arm is to be used for machining, the movement path of the arm may be very complicated, and even some obstacles may be evaded, which makes the control programming of the arm more difficult. The Siemens control system can be automatically inserted. Supplementary paths and other advanced features to simplify control programming problems.
Overall, Siemens believes that the combination of the robotic arm and other field devices will be the trend of future arm applications, and this trend will bring some competitive pressure on other existing field devices. In addition to machine tools, An AGV with an arm may also replace the application requirements of a part of the conveyor belt.
Taiwan has ranked sixth in the global robot market for six consecutive years. Immediately after Germany, it is the key development market for automation. General Managers Northeast Asia General Manager Shan Gengang shared that Universal Robots expects global revenues in 2018. Growing at least 50%, and the market revenue of Taiwan's Northeast Asia region is expected to be higher than the global growth rate of more than 50%. Yamagen said that it is not only the traditional and processing industries, but also other industries in Taiwan. Such as retail, logistics, agriculture, medical care is also an industry with great potential to develop automation.
3D modeling improves manufacturing efficiency. Aerospace production speed is increased by five times.
In the process of transforming intelligent manufacturing, management processes and design and manufacturing methods must be fully upgraded. Hanxiang Airlines has successfully improved design and production efficiency and reduced production costs through intelligent manufacturing layout and 3D digital simulation.
Lin Nanzhu, general manager of Hanxiang Airlines (pictured left), said that in the face of today's customers' requirements for production speed, it must be upgraded to four times that of traditional manufacturing to meet the requirements. To fully improve production efficiency, it must be managed by the system. , design, manufacturing and development of all aspects of digital transformation. At present, Hanxiang Airlines has reached nearly 90% of the digital proportion, and is expected to save 20% of the cost through intelligent manufacturing by the end of 2018.
Figure 5 Li Zhijun (right), vice president of Value Solution Greater China, said that in the future, Dassault hopes to have partners in all walks of life. The left is Lin Nanzhu, general manager of Hanxiang Airlines.
Lin Nanzhu said that smart manufacturing brings many changes and challenges to the aerospace industry. Digital transformation can help simplify production processes, increase production efficiency, and enhance corporate competitiveness. Therefore, Hanxiang Airlines continues to actively promote digital transformation and intelligent manufacturing. Achieve the intelligent management mode of the process.
Hanxiang Airlines is an important supplier of Airbus SAS and Boeing. Lin Nanzhu further explained that in the face of global competition, if it does not have digital capabilities, it may not be the first choice of international companies. In the face of upstream suppliers, Xianghang Airlines also has requirements for its digitization. However, the speed of domestic suppliers’ transformation is not as fast as expected, mainly considering the issue of security and its digital transformation strategy. More conservative.
In the process of transforming intelligent manufacturing of Hanxiang Airlines, the various platforms provided by Dassault System have helped many. Lin Nanzhuo pointed out that Hanxiang can use 3D digital simulation to find the platform provided by Dassault. Design shortcomings, and then improve the design.
Li Zhijun, vice president of Value Solution Greater China, Dassault System pointed out (Figure 5 right), the company has cooperated with Hanxiang Airlines for more than 20 years, and both sides' scientific and technical experience will be communicated at any time. In the future, Dassault's goal is Can have partners in all walks of life, and assist partners to train and enhance the degree of digitization to provide accurate value-added services to their customers.