What isa robotcontrol system?
Here we present the scope of a robot control system using an analogy and referring to the hierarchical form of the computer industry.
Why is the control system of a robot crucial?
Like the operating system of a computer, the robot control system plays the role of the most basic software platform in the entire industrial robot industry.
When you buy an industrial robot, the basic functions that come with the control system can be programmed to meet the needs of many applications. If the functions do not meet the requirements, then you can choose to install some industry-specific application software packages, such as grinding, force control, sorting, welding, spraying and so on. This is like simple image processing we can use Window's own drawing software or even PowerPoint to realize, but if the requirements are higher, then you can install an independent application software like Photoshop.
The control system provides the platform, the application software on the basis of the platform to expand the scope of application and enhance the use of experience.
Throughout the history of industrial robotics, there have been many excellent niche application software companies that have successfully expanded the range of robot applications in a particular niche direction by using new technologies and researching new processes. Some of these companies have grown independently to become application experts in a certain process area, using special robots customized with robot manufacturers plus their own unique application software packages in the niche areas of leadership, such as spraying DURR, welding CLOOS, etc.; some have been successfully acquired by the robot manufacturer, and its application technology into the framework of the robot control system.
In the past few years, deep learning technology has many application cases in robot path planning and vision. These technologies have enhanced the application level of robots when facing unstructured environments, and also solved some application problems that could not be solved by fixed programming before, which is of positive significance for expanding the scope of application of robots. As a robot manufacturer, we believe that new application technology is an important driving force for the development of the robot industry, and we are willing to grow together with reliable partners.
So which is more difficult and has a higher threshold, making a system or making an application?
There is no absolute answer to this question, depending on the depth and breadth of the do. Microsoft does not dare to say that it will be able to make image processing software comparable to Photoshop, Adobe may not be able to do Windows with full commitment for decades, the final answer depends on the company's development history and strategic trade-offs, and I believe that every professional who is deeply engaged in the industry will have their own judgment on this.
However, if you deliberately confuse these two different levels of concepts, deliberately belittling others to elevate themselves, it is either a lack of correct basic understanding of the industry, or in the claptrap. Theoretical maturity does not mean the threshold is low, nor does it mean that the products and industry are mature. Everyone understands the principle of the internal combustion engine, and you make a V8 out of it to see?
The control system has the lowest threshold? Wrong!
Robot by definition is "software-driven mechanical system": similar to computers and cell phones, industrial robots are a very typical combination of hardware and software, electromechanical integration products.
Such products follow the principle that the hardware determines the performance boundaries and the software performs the hardware and defines the behavior of the product (robot).
In relatively mature markets, when the hardware used tends to be the same, it is common practice to differentiate and create value through software. Industrial robotics industry after decades of development, hardware progress has slowed down greatly, the mainstream manufacturers of hardware configuration is basically the same, domestic manufacturers can also buy with imported brands of similar spare parts (the current price will be much higher), then the robot control system becomes the main factor in determining the performance and functionality of the robot.
Suppose we have two robots with the following configurations:
1. imported abb's robot controller + domestic servo system + domestic reducer + domestic mechanical body;
2. Domestic robot controller + imported servo system + imported reducer + imported mechanical body;
So which one will have stronger performance and more functions?
The answer is easy to see, no one would think that the second robot equipped with a domestic controller can be PK over the first robot using abb robot controller, even if the performance of the program in addition to the controller is almost all the use of domestic parts.
Since the control system is so important, and so far no domestic development can be close to the market mainstream robotics brand native system performance controller products, then why there will be "control system threshold lowest" conclusion, or illusion? I think there are three main reasons:
1. The lack of domestic robot control and machine product development has in-depth knowledge of the team and the company. The correct judgment of an issue depends on an in-depth understanding of the problem and the background. Domestic industrial robotics industry started late, the lack of high-performance robotics product research and development of the team and the company has enough experience accumulated, the mistaken belief that the robot as long as it can move then the control system is enough, resulting in many people wrongly believe that the control system is very simple. In the latter article, we will discuss some details about the development of modern robot control system.
2. Domestic robot products are still generally oriented to the low-end market, customers have almost no requirements for the performance of domestic robots. In this market demand, many companies on their own production of robotic products can only be required to move, to meet some of the precision, speed, functional requirements of the occasion is not high, then the use of third-party control systems or the use of a variety of motion control cards cobbled together the controller is able to meet the requirements. We can not say that this approach is wrong, because the market demand determines the product positioning and design, since there is a large number of low-end demand in China, then such a robot product also has its value of existence. But in the relatively limited market space in the field of industrial robots, long-term low-end-oriented companies will not be able to get enough profit to support its move to the middle and high end, if all the domestic manufacturers are like this, then the domestic robotics industry will most likely repeat the mistakes of the machine tool industry, in a short period of sales climax, fell back to a large number of shipments but most of the profits were taken away from the foreign CNC software system vendors in the situation.
3. People can not see and can not directly compare the lack of intuitive and accurate understanding of the parameters. Like the reducer such as visible and tangible parts, experienced teachers can be very intuitive to see the domestic products and imported products and how big the gap between. The public is more accustomed to focusing on the visible gap, and ignore the real hidden behind the gap. In fact, this phenomenon in the computer industry can also be referred to as an example, you can now think back to their usual productivity software (Office, CAD, CAE, scientific computing, simulation and analysis, video editing, audio editing, graphic design, etc.), how many of them are domestic manufacturers to provide? Almost none. Genuine domestic users need to pay a huge amount of money every year to buy these software products. But how long ago do you remember hearing that there is a big gap between domestic and foreign CPUs? How long ago did you hear someone talk about the large gap between Chinese and foreign software? Or have you never heard of a big software gap? The gap between Chinese and foreign productivity software products may not be as large as that between chips, but still can not hide the backwardness of the domestic professional software, and it is the spread of this thinking has led to the robot control software is not valued.
From 1973 ASEA (1988, ASEA and BBC Brown Boveri merged to form the current ABB) released the world's first all-electric industrial robots to now more than 40 years have passed, during which time, although the speed reducer and servo system has been in the continuous reduction of the volume, improve the responsiveness, reduce transmission errors and other improvements, but this improvement more To be precise, it is a gradual improvement rather than a breakthrough. Industrial robot applications can be used from the initial assembly of gearboxes to the current welding, painting, assembly, patch, dispensing, grinding, polishing, sorting, handling, palletizing, testing, injection molding, cutting, measuring, stamping, packaging and other dozens of fields, the vast majority of which rely on robot control systems in the architecture, control, planning, process, human-machine interaction and other aspects of the innovation.
Take the recent very hot collaborative robots as an example, although it has a very different appearance from traditional robots, but it should still be regarded as a product innovation based on traditional motors and speed reducers, or more accurately said to be the innovation of control and software innovation.
Out of the "three core components" of the misunderstanding
Control system, servo system, reducer is known as industrial robots, "three core components" argument has a long history, but rummage through the Google also did not find this argument is when and where the rise, but often used as a starting point for argument, it seems that each robot is to buy the three main components, and then design the mechanical It seems that every robot is made by buying three main components, designing the body and the control cabinet, and then producing it.
This is fine if you are talking to someone outside the industry for a quick explanation of robot construction. But as industry insiders, especially the robot manufacturer, if you have been in accordance with the main components to save the robot thinking, it is difficult to make excellent products.
Now the situation of domestic robots is like the domestic cell phone five or six years ago. Relying on the chip manufacturers to provide solutions, a small workshop can quickly save a cell phone, such products in a not yet saturated + not yet enlightened market may have a place, but when the blue sea into a red sea, the first to be eliminated will be the shoddy cottage goods, and then there is no core technology of the second-rate brands that are able to hold out to the end of the several domestic giants. So successful cell phone manufacturers may also have initially piled up the history of hardware, but the end result is inevitably through excellent machine design and good software adaptation to find their own development path.
Imagine a number of years from now, the robot market, the winner, relying on the market will not be able to buy the so-called "core components", but has the core algorithms and perfect function of the control system software as well as excellent machine product development capabilities. Because the robot is first of all "software-driven mechanical system", is a software-driven electromechanical system; as the system as a whole, but also as a whole and design. The reality is exactly the same: mainstream robots have their own control systems. The hardware determines the performance boundaries, the software performs the hardware and defines the behavior of the robot. Any manufacturer that wants to actually build a robot will not pass up the opportunity to define the behavior of their own robot.