The hottest industrial robot intelligent developme

  • Detail

Intelligent development of industrial robots: vision and tactile sensing simplify robot tasks

the robotindustriesassociation points out that from 2003 to 2005, the robot industry in North America developed at an average annual growth rate of 20%. Under the background of weak demand in the automobile market and increasing pressure from foreign manufacturers, how did this success be achieved? The general reduction of costs is one of the reasons. From robots to complete systems, prices are constantly falling. The continuous improvement of robot performance is also one of the reasons driving this strong growth. Robots today can perform many tasks that could not be performed a few years ago. Robots can also accomplish more things in a shorter time, providing a higher level of productivity

the most important reason should be the continuous improvement of robot intelligence. From the advent of robots to the present, there have been several levels of development in intelligence in response to customer problems within two hours of deciding on components, obtaining, checking the positioning of their own components, and detecting error states or related factors. Such intelligence is mostly based on specific sensors used to detect a specific state

for example, the robot will use a "photoelectric eye" to determine whether a chip pin, brake or other components are ready and in the right direction. During the development of electronic products, this "photoelectric eye" is connected to PLC or directly to the robot controller. At the right time, the robot will check the "photoelectric eye" to ensure that a part is in place and in the right direction, and then it will pick it up or carry out other operations

using "photoeye" or similar sensors is a simple and reliable way, and it is also the right choice. However, automation is not always that simple. There are various types of components to deal with, so the method of distinguishing components may be more complex

no matter the part or the manufacturing process, a simple transmitter cannot be applied. For example, parts placed in cabinets separated by sliding pallets are generally used for metal parts. The components with complex geometric structures are difficult to locate accurately without additional devices

two-dimensional visual positioning

for a long time, due to the inability to find a suitable method to deal with these complex matters, the use of robots in some industries has always been unable to expand. It will overcome some problems that cannot be solved by traditional foaming materials. Recently, this situation has finally changed. The most important technology is the two-dimensional vision system. Please log in to the automatedimagingassociation site to get more information about this technology

2d vision system is composed of standard industry cameras, which are used to take photos for robots to analyze and decide which parts to deal with. The history of industrial vision system is not long, but now it has developed to be used in some applications in terms of price, performance and reliability, which could not be realized just a few years ago

using the vision system can assist the robot to pick up the parts placed in the cabinets separated by several layers of standard sliding pallets. This is a general method of transferring parts from one workshop to another or even within the same workshop. If there is no visual system, the manufacturer must use relatively expensive formed plastic padding or other methods to accurately extract the components in the box. This kind of formed plastic padding that can be placed in the cabinet is relatively more expensive, among which the design and manufacture of the mold alone requires us $60000 to US $100000

2d vision system can well replace the formed padding or other more expensive ways to locate the components in the cabinet. However, until recent years, there are still some reasons that make the use of the visual system difficult, including the differences in the colors of different batches of parts, the differences in the specific conditions of cabinets, and the marks on the reused partitions, etc. With the continuous development of vision technology, these problems can now be well solved. Today's 2D vision system can locate most of the components stacked on the top of the partition inside the cabinet

the general method of this application is to use a camera installed on the cabinet to locate components. The camera is installed at a sufficient height, so the robot can move under the camera and enter the cabinet. At the beginning of each layer, the robot will analyze the pictures of the components of this layer sent by the camera, and then locate them one by one

if you need more accurate positioning, you can also install the camera on the robot, and the robot will move the camera to the upper end of one or a group of components and take photos, and then send the photos back to the robot for analysis and positioning. When all the components on a certain layer are removed, the robot will remove the partition and begin to remove the components on the next layer

whether the camera is fixed or installed on the robot, the increased cost is much lower than using a special pad or other methods to locate the parts. The vision system can also bring more flexibility to deal with different parts on the same line, or to adapt to the replacement of parts. Due to cost reasons, such automatic stacking and destacking could not be realized a few years ago. Now, with the vision system, the robot scheme is feasible and affordable

another benefit of the vision system for robot use is that it reduces the cost of the transmitter used to present components to the robot. Before using a 2D vision system, many components must be positioned on a fixed pallet conveyed by a pallet conveyor. Even a simple pallet conveyor costs $30000, and the cost will increase with the increase of the transmission distance of the conveyor. With 2D vision technology, components can be transported on relatively inexpensive belt conveyors. Parts placed on the conveyor by the operator or robot will then be transferred to the robot. When a camera installed at the end of the conveyor detects that a part has been transferred to the end, the conveyor will stop immediately. The camera will also locate the parts to assist the robot in picking the parts with low modulus

as used to remove parts from the partition in the cabinet, the vision system on the conveyor belt can also assist the robot to accurately locate and remove most parts

Copyright © 2011 JIN SHI