Robotics and artificial intelligence have long been gaining ground in our everyday lives. Logistics is no exception, and industry-compatible robotics solutions are gaining importance in the sector. The days are gone in which robotics systems were considered extravagant toys or costly investments. This is a good thing because we know: It works!
Equipped with artificial intelligence from the AI start-up Covariant, we have developed an industry-compatible solution with our Pick-it-Easy Robot – a fully automatic commissioning robot that has no equal in logistics.
But how does the Pick-it-Easy Robot learn? Why are KNAPP and Covariant the perfect match when it comes to the development of robotics solutions that can be successfully used in everyday industrial applications? And the most important question for us is: How happy are our customers with the Pick-it-Easy Robot’s performance? To answer these questions and introduce you to the exciting world of our robot’s artificial intelligence, we caught up with our experts.
Insights from our experts
Peter Puchwein, Vice President of Innovation, is convinced: Innovation is a vital part of our company DNA. It’s a motto that Peter and his team live by. Considerable expertise combined with a great deal of passion and fascination for robotics has enabled the team to take the AI of the Pick-it-Easy Robot to a whole a new level.
How would you explain artificial intelligence? How does the Pick-it-Easy Robot learn?
It may seem funny, but you can compare artificial intelligence with raising a child. We reward kids for their good habits and help them give up bad ones. Artificial intelligence is no different: If the Pick-it-Easy Robot picks a new product correctly at the first try, we reward it with bonus points. The better the pick, the more bonus points it gets. The system learns non-stop and autonomously. Improvements are automatically integrated into the basic model and distributed to all robots. The Pick-it-Easy Robot does not only learn to improve itself, but also to improve its bionic “siblings”.
Peter Puchwein: Transferred to the real world, you would only have to educate one child to automatically achieve the same effect in the siblings. When one child makes a mistake and learns from it, the siblings won’t make it in future. In the virtual world, we have access to the brains of all Pick-it-Easy Robots and can improve their content with software updates. This is exactly the added value we offer to our customers. Whenever one robot acquires new knowledge, it is immediately available in all the other robot stations.
What is the story behind your partnership with Covariant?
Peter Puchwein: We formed our partnership with Covariant after our Robotics and Machine Learning teams participated in the GTC – an important AI conference in Silicon Valley – at which Pieter Abbeel, the co-founder of Covariant, gave a talk. From the get-go, the chemistry was right both content-wise and on a personal level. We decided to conduct some tests together to challenge Covariant’s artificial intelligence. The results were astonishing, so we decided to become partners and combine our expertise in robotics and logistics with Covariant’s AI brain.
Ted Stinson, COO, Covariant, is also delighted about the partnership with KNAPP:
Ted Stinson: Our vision was to build a cognitive system for a robot, so we asked ourselves, “Where are the opportunities to bring this sort of AI robotic capability to the market?” Along the way, we met our now very good friends and partners at KNAPP and understood the enormous potential a partnership between the two companies would have. At first, it was a process of discovery, of trying to see: Is it possible to have the core strengths of Covariant and the capabilities of KNAPP come together in a set of solutions?
Today we couldn’t be more thrilled about the partnership with KNAPP and are very excited about the first product we have brought to the market together – the new generation of the Pick-it-Easy Robot.
Ted Stinson: If you think about the modern warehouse, KNAPP has done amazing work. Pretty much everything that could be automated to prevent people from trudging long distances in the warehouse has been automated thanks to KNAPP’s ideas and pioneering innovations of the last couple of decades. What remains to be solved are the precise tasks of the human hand.
And this is where we come in: The essence of the intelligent Covariant software is a system that learns. Just think of the modern warehouse. There are tens of thousands, hundreds of thousands, sometimes even over a million different products, all of which are changing on a very regular basis. The robot has to learn rapidly and accurately.
Pick-it-Easy Robot in action
We are proud that the newest generation of the Pick-it-Easy Robot is already delivering a successful performance at our customer Obeta. Obeta is a German wholesaler of electric devices with headquarters in Berlin and boasts 28,000 customers and 64 locations across Germany. With the help of the Pick-it-Easy Robot, they pack thousands of customer orders every day. Michael Pultke, Head of Logistics, Obeta, provides some insights:
What tasks does the Pick-it-Easy Robot perform at Obeta?
Michael Pultke: We pick a diverse range of products from our OSR Shuttle ™, which has 78,000 storage locations for small load carriers. The products come in different packaging and quantities and have to be carefully gripped and packed into dispatch containers or cardboard boxes, both quickly and continually. The Pick-it-Easy Robot takes the load off the manual picking stations by handling the routine picks. Our goal is to make the picking process as smooth and efficient as possible.
In your view, what added value does the Pick-it-Easy Robot offer?
It’s definitely the seamless operation and continuous monitoring of storage processes that represent the added value. Automating our warehouse operation is one of Obeta’s top priorities and the Pick-it-Easy Robot helps us achieve this goal.
Further reading recommendations
E-Commerce brachte in den letzten Jahrzehnten eine der größten und weitreichendsten Veränderungen in verschiedensten Branchen mit sich, vor allem in der Fashion-Industrie, denn in der Entwicklung des Online-Handels ist die Fashion-Branche Spitzenreiter. Was dieser Anstieg für die Logistik bedeutet und wie Robotik hilft, Herausforderungen zu meistern, lesen Sie hier.
Robots are already supporting workers in countless logistics areas. In the food retail sector, it’s KNAPP’s Pick-it-Easy Robot who is playing a pivotal role in meticulously processing the diverse range of articles.
Machine learning is everywhere. But what exactly lies behind this future technology? Is it really possible to teach machines to learn? What advantages does it offer– and where does KNAPP come into play? Let’s take a closer look.