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.

近年来哪些技术发展对散乱零件抓取技术产生了重大影响？

Arquemino Lopes Junior：在我们看来，主要有三项技术发展改变了散乱零件抓取：人工智能和机器学习的影响、图像处理和物体识别的进展，以及实时数据的使用。这些技术显著提高了系统的精确性和灵活性。

最重要的目标仍然是计算出最为合适的抓取点，并将“手臂端”的抓取系统设计得尽可能高效。为了将抓取错误降到最低程度，Festo 将气动、电气和数字自动化的所有优势应用到寻找解决方案中。

我们抓手技术产品系列种类丰富，能够涵盖大量应用场合。特别要强调的是我们的定制解决方案。这些方案通过与客户进行有针对性地合作而开发得来，用以开放适应个性化需求的抓手解决方案。在我们发起的“Grip it”开发倡议框架下，一个跨国团队通力合作，专门致力于开发特殊抓手。

如何利用自动化过程中的数据来优化智能仓储效率？

Arquemino Lopes Junior：一个很好的例子是预测性维护 (Predictive Maintenance)。许多客户使用我们的创新软件 Festo AX 处理其机器设备中的实时数据，并通过人工智能进行分析。这样将停机时间减少高达 25%，并持续提升生产效率。在电气自动化领域，我们也同样依赖 Festo AX。通过分析电机的电力消耗，我们能够了解电机一天中的工作情况。能耗峰值通常表明转矩需求增加以及存在机械问题。通过 Festo Ax，我们能及早发现性能下降的原因，防止潜在故障，避免产生问题。实时数据还使我们能够清楚了解，所有组件是否合适，其尺寸设计是否正确。这样能够提高效率、将磨损减小到最低程度、延长设备使用寿命，并降低运营成本。

Festo 如何通过 GripperAI 支持现代智能仓储的要求？

Arquemino Lopes Junior：GripperAI 绝对是一项针对散乱零件抓取的创新。通过这项革命性软件解决方案，我们能够使机器人和搬运系统精确灵活地抓取不同形状和尺寸的无序物体，即使它们的摆放处于混乱状态。我们的解决方案利用人工智能，尤其是机器学习，来自动适应不断变化的抓取任务，持续优化抓取过程中的效率。因此我们不再需要 CAD 模板来配置摄像头或进行耗时的机器人示教。GripperAI 能够自主识别物体，并根据实际情况计算最佳抓取点。这增加了解决方案的灵活性，并大大减少了手工操作量。

测试中，GripperAI 在简单任务中几乎达到了 100% 的成功率。即使在物体无序、环境混乱的复杂场景中，成功率都同样很高。这表明软件能动态地适应不同的要求。

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.

What would you recommend to companies that want to integrate bin picking into their processes?

Arquemino Lopes Junior: Make sure you choose a partner with broad automation expertise and an extensive portfolio in gripper technology. This is the only way to meet solutions that fulfil the specific requirements of the application. Standard solutions often reach their limits, especially in the area of bin picking. The individual requirements and circumstances are so diverse in many areas of application that customised approaches are essential. Your partner should therefore also be prepared to develop individual solutions together with you.

Since many machines are often used in different countries, it is essential that the automation partner also has a worldwide presence. This not only guarantees the availability of spare parts, but also the smooth operation of the systems.

And probably one of the most important recommendations is to choose a partner who develops across technologies and offers everything from a single source, be it pneumatic, electric or digital solutions. Which technology is used always depends on the requirements of the application in question. A partner with expertise in all areas is best placed to help you develop the right automation strategy for your processes.