Robotics In Factories

Share on facebook
Share on twitter
Share on pinterest

Introduction to Remote Sensing:-

 In the past, factory production lines were automated for mass production, and many industrial robots and specialized machines were introduced. Recently, flexible manufacturing systems, such as the cell production system (unlike in the line production method, an entire product is assembled by one worker), are being introduced in an increasing number of production sites in order to deal with differentiation of products and to meet diversified needs. 

However, many of the tasks in flexible manufacturing systems rely heavily on workers because the number of parts to be handled is larger so the time and costs required to switch product types on robots and specialized machines is greater. 

Recently, because of the decrease in the working population due to Japan’s aging society with a falling birth rate, there are expectations that tasks which rely heavily on workers will be automated by using industrial robots in combination with sensing technology and production know-how. With “intelligence” as their motto, industrial equipment manufacturers are focusing on developing technology that will automate tasks that are currently performed by humans, but the types of tasks that have so far been automated are very limited. In order to automate factories of IHI group companies and /or customers (present and potential), IHI is also conducting research and development for the development of a new system that combines IHI’s hard-earned innovations, such as

(1) advanced sensing technology

(2) control technology,

(3) mechanics technology with industrial robots, for the purpose of making it possible for these robots to perform tasks that have yet to be automated and that rely heavily on human workers. 

This research introduces the concept of a system in which industrial robots are applied to

 (a) Picking work

(b) Medium payload handling work,

 (c) Assembly work and also introduces the research and development of such a system.

2. Bin picking robotics using 3D object recognition

 At production sites, parts in bins are arranged and transferred by workers. Parts such as bolts are stored in bins, and it is easy for humans to handle several types of parts at a time. However, it is extremely difficult for robots to measure parts in bins with a sensor, recognize individual parts, and take them out. IHI is developing bin picking robotics using its own proprietary 3D object recognition technology (1). The main feature of this system is that unlike conventional systems that use cameras (2.5-dimensional or stereo vision), this system is not affected by ambient light because it adopts lasers as the three-dimensional measurement method. In addition, this system is highly compatible with 3D CAD systems and facilitates switching product types when used with 3D CAD data, providing enhanced usability.

3. Medium payload handling

Currently, medium-weight objects (50 to 200kg) are handled mainly by workers or by operating a crane. Many production sites are expected to be automated because automation provides improved safety and productivity. However, because objects to be handled are generally large and heavy, it is difficult to measure and position them, causing a delay in automation with robots. IHI developed a system that can load and unload engine blocks, etc. of around 60 kg onto and off of pallets, and automatically transfer them to the inspection process when necessary. 

The combination of the hand eye method, in which a camera is installed in the robot’s hand and 3D object recognition technology, allows a robot to recognize work pieces even if they are not perfectly aligned or if the pallets onto which the work pieces are loaded have a manufacturing tolerance. In addition, the adoption of a hand tool equipped with a floating mechanism for grasping a work piece from above enables smooth unloading of work pieces from pallets and palletizing (loading of work pieces onto pallets) even if work pieces are tightly arranged in a pallet.

4. Hand guiding system

 Assembly and handling of work pieces with complicated shapes requires precise positioning. To fully automate such tasks, expensive sensors and advanced and complicated controls are needed. In addition, there are still many problems to solve, for example, it may be impossible to measure some parts of a work piece to be positioned depending on its shape, and even if automation is successful, “minor stoppage” (equipment does not fail but temporarily stops due to minor abnormalities, though it can be restored in a short time) occurs frequently, preventing the utilization ratio from increasing. 

One possible solution to address these problems is to classify tasks into those that robots are suited for and those that humans are suited for so that robots and humans can work cooperatively. IHI developed a hand guiding system that complies with ISO 10218-1: 2006 in order to enable cooperative operations between industrial robots and humans.

This system can be operated in manual and automatic modes, and the robot performs tasks that can be performed based on teaching-playback and sensing. The worker, while being assisted by the robot, performs tasks that cannot be performed based on teaching or sensing, offering the advantages of using industrial robots as more than a mere power assist system. In order to apply the system to a wider range of tasks, IHI will perform various tests to evaluate and improve this hand guiding system.

5. Cell production robots

 In order to meet diversified market needs, the cell production system has been adopted in various fields because it is suitable for flexible manufacturing systems where various types of products are manufactured according to demand. At the same time, it has become difficult to secure workers at production sites, and therefore, automation that utilizes the advantages of the cell production system is increasingly required. However, because the cell production system requires many more types of tasks and parts, there is a need to make design changes or prepare special jigs for automation, and the tact time becomes longer, making it difficult to take advantage of the benefits of automation. 

IHI is developing assembly robots by making robots more intelligent so that they can be made to function in multi-product assembly work that has yet to be automated. These assembly robots can operate in the same cell production system work environments as human workers. Therefore, the existing workbenches, jigs, and dedicated automatic equipment currently used by workers can be used as is.

The intelligence technology for flexible manufacturing systems that this prototype employs is a force control that enables assembly with an accuracy of a few micrometers and a visual sensor that enables the robot to correct misaligned parts and screw holes. This technology has eliminated the need for dedicated aligners and jigs used for the accurate positioning of parts. In addition, the tact time is almost the same as that needed for workers.

7. Conclusion

This research has introduced the concept of a system in which industrial robots were applied to picking work, medium payload handling work, assembly work, etc. by combining IHI’s hard-earned innovations, such as advanced sensing technology, control technology, and mechanics technology with industrial robots, and has also introduced the research and development of such a system.

 It is becoming possible to apply industrial robots to tasks that cannot easily be automated and thus rely heavily on human workers. In addition, robots work long hours and handle heavy objects without getting tired or making mistakes, leading to improved quality.


(1) T. Hayashi, M. Sonehara, T. Inoue, T. Shima and Y. Kawano : Development of Bin Picking Robotics using 3-D Object Recognition, Journal of IHI Technologies Vol. 48 No. 1 pp.7-11 

(2) M. Fujii, D. Shiokata, H. Murakami and M. Sonehara : Proposal of Safety System for Human and Industrial Robot Collaborative Working, ROBOMEC 2008 Proceedings (2008.6) 2A1-A21