Why is bin picking so important in intralogistics?

Arquemino Lopes Junior: The increasing variety of products, the ongoing boom in e-commerce and the growing complexity of logistics processes require intelligent automation strategies. Companies need to work more efficiently and cost-effectively in order to remain competitive in this environment. Bin picking is the automated part of intralogistics. It enables companies to remove a wide variety of objects precisely and reliably from chaotic or organised containers. With this technology the flow of materials can be optimised and bottlenecks in the logistics chain can be avoided.

What challenges, opportunities and risks do you see in this context?

Arquemino Lopes Junior: The big challenge is to develop gripper systems that can flexibly and reliably grip a variety of different items. A system that can really grip any object is like searching for the Holy Grail. The variety of product geometries and materials poses an enormous challenge for automated gripping. If the bin-picking process is not frictionless, it quickly becomes a bottleneck in intralogistics. The risk is that the material flow comes to a standstill and the performance of the overall system is reduced. On the positive side, companies that successfully integrate intelligent bin picking into their processes are taking a decisive step in the right direction. They not only boost their efficiency, but also create the basis for surviving in a market that demands ever faster and more flexible solutions.

In your view, which primary drivers in the market are responsible for this development?

Arquemino Lopes Junior: Online retail is certainly one of the biggest drivers of the current challenges in intralogistics. The pandemic has given a massive boost to e-commerce, and although growth has levelled off somewhat since then, the trend remains positive and sustained. This means that the steadily increasing number of orders is also generating a significantly higher processing requirement in logistics. In addition to this development, the variety of products is constantly growing, which means that more different product geometries and materials have to be processed in the bin picking process.

Our expectations as consumers also play an important role. We expect the products we have ordered to arrive quickly, preferably on the same day. This expectation presents companies with the additional challenge of processing the already high volumes in e-commerce even faster. And this shows very clearly that it is crucial to have smart, automated bin picking to ensure efficiency, speed and reliability.

What technological developments have had a particular impact on bin picking in recent years?

Arquemino Lopes Junior: In our view, three technological developments in particular have proven to be game changers for bin picking: the influence of AI and machine learning, advances in image processing and object recognition, and the use of real-time data. They have significantly improved the precision and flexibility of the systems.

The most important goal is still to calculate the perfect gripping points and develop the gripper system at the so-called “end of arm” as efficiently as possible. At Festo, we combine all our expertise in pneumatic, electric and digital automation to find solutions that will reduce errors to a minimum.

With our broad portfolio of gripper technologies, we can cover almost any application. I would also like to highlight our Customised Solutions, which are the result of specific collaboration with our customers in order to develop customised gripper solutions. As part of our development initiative “Grip it”, a transnational team of people who are specialised in developing special grippers are working together on this.

How can data from automated processes be used to optimise efficiency in intralogistics?

Arquemino Lopes Junior: A good example is predictive maintenance. Many of our customers use our innovative Festo AX software to process real-time data from their systems and machines and analyse it using artificial intelligence; this enables them to reduce downtimes by up to 25% and sustainably increase the efficiency of their production. But we also rely on Festo AX in electric automation. By analysing the power consumption of an electric motor, we can see how it has operated throughout the day. Peaks in energy consumption often indicate an increased torque requirement and mechanical problems. Using Festo AX, we identify the causes of such performance drops at an early stage and prevent potential faults before they become a problem. Real-time data also gives us valuable insights into whether all components have been optimally selected and dimensioned; this increases efficiency, minimises wear, extends the service life of the systems and reduces operating costs.

How does GripperAI from Festo support the requirements of modern intralogistics?

Arquemino Lopes Junior: GripperAI is an absolute innovation for bin picking. This revolutionary software solution enables robots and handling systems to grip disorganised objects of different shapes and sizes precisely and flexibly, even if they are randomly positioned. Our solution uses artificial intelligence, in particular machine learning, to adapt autonomously to changing picking tasks and continuously optimise efficiency in the picking process. This means there is no longer a need for CAD templates to configure the camera or time-consuming teaching of the robot. GripperAI recognises objects independently and calculates the best gripping points depending on the situation. This makes the solution extremely flexible and significantly reduces manual effort.

During testing, GripperAI achieved a success rate of almost 100 percent for simple tasks. Even in complex scenarios with disorganised objects and chaotic environments, the success rates were similarly high. This shows how dynamically the software can adapt to different requirements.

Where is GripperAI already being used and what further developments are planned?

Arquemino Lopes Junior: We are currently using GripperAI in various pilot and research projects to test new functions and applications. By working together with logistics companies, we can trial the technology for gripping different parts, from small objects such as USB memory sticks or cans to heavy boxes.

Planned enhancements are aimed at further increasing the software's learning ability and adaptability; this also includes the integration of additional sensor and image processing technologies to further optimise the precision and efficiency of the gripping processes. One particular highlight is our FLAIROP research project. As part of this project, we are investigating how grippers can become even more flexible and intelligent through the use of integrated cameras and distributed AI. Our aim is to use training data from all locations without disclosing sensitive company data.

您会给那些想要将散乱零件抓取集成到生产流程中的企业什么建议？

Arquemino Lopes Junior：务必选择一个具有广泛自动化专业知识并拥有丰富抓手技术产品组合的合作伙伴。只有这样，才能开发出满足特殊应用需求的解决方案。而标准解决方案在散乱零件抓取领域往往有其局限性。在许多应用领域有着各种不同的个性化要求和条件，定制解决方案是必须品。因此，您的合作伙伴应该愿意与您共同开发个性化解决方案。

由于许多机器常常在不同的国家使用，选择一个全球范围内布局的自动化合作伙伴至关重要。这不仅保证备件的供应，还能确保系统的顺畅运行。

也许最重要的建议是选择一个跨技术开发并提供一揽子解决方案的合作伙伴，他能够提供气动、电气以及数字化解决方案。采用哪种技术始终取决于相应应用的要求。一个在各个领域都具有专业能力的合作伙伴，能够帮助您根据自身工业流程打造自动化战略。