Product Information
Warehouse Robotics for SAP-Extended Warehouse Management
As of the late 1980s, heavy-duty industrial robots have been implemented by large system integrators to automate the manufacturing lines of large companies. They pay off at scale, but the huge investments and technical know-how necessary to install or operate them, created a significant barrier of entry.
Today, a new generation of robots, so called collaborative robots, makes warehouse automation affordable and useful for even one-off tasks. Yet challenges remain:
- High upfront costs: System integration is bespoke, time-and-material work, mostly in the hands of OEM or system integrators, that results in expensive and inflexible solutions
- Vendor lock-in: Vertically-integrated offerings provide more generalized solutions with high dependency but limited flexibility
- Silos: Multiple solutions are needed to cover use cases, but inter-operability is limited, and tooling is disjoint
- Inconsistent IT Strategy and Governance: Unclear security and management architecture across sites, de-centralized proprietary monitoring and management tools
To mitigate these challenges, SAP now released SAP Warehouse Robotics. It comes as an add-on to Extended Warehouse Management (SAP-EWM) with out-of-the-box EWM scenarios. This new solution dramatically reduces the need for specialized system integrators, promotes re-use, and ensures compliant IT-security and management via:
- OEM agnostic architecture – maximum flexibility, freedom of choice
- Multi-device and multi-use-case fleets – business demand driven roll-out (automation on demand)
- Out-of-the-box EWM integration with predefined scenarios
- IT Governance which automatically includes robots
SAP Warehouse Robotics works with warehouse robots of most vendors with very little adaptation and can run heterogeneous fleets. Users of the solution gain strategic flexibility and a low-cost entry into warehouse robotics:
- They can expand into areas where robots haven’t been used. This applies to non-repetitive tasks and assistive tasks. It also applies more generally to small and medium-sized companies, who could not afford robots in the past.
- They can scale ad-hoc and, thus, optimize machine usage
- They avoid getting trapped into a vendor lock-in and can run fleets from multiple vendors without the hassle of managing a variety of software stacks.
Last, but not least, the solution is provided as Open Source. Anybody can download it for free and use it or make modifications. The Google part of the Code is here and the SAP part is here.
This is the first blog of a new series that will provide an introduction to Cloud Robotics. Stay tuned for more!
Hi Albrecht,
Thank you very much for Sharing.
"Last, but not least, the solution is provided as Open Source. Anybody can download it for free and use it or make modifications"
Is it the correct below link ? Please help.
https://github.com/SAP/ewm-cloud-robotics
Kind Regards,
Ajit
Hi Ajit,
Thanks for pointing this out! I just added the URLs into the blog. Also, your URL is correct.
Kind regards,
-Albrecht
Hi Albrecht,
What is the role/benefit of using SAP Cloud Platform in between?
I feel like you couldn't solve something with SAP EWM and used SAP Cloud Platform, what is that?
Best Regards
Serhan
Hi Serhan,
The SAP Cloud Platform enables the communication between SAP systems (EWM in this case) inside the company's firewall and the outside world.
Technically, you can do without if you are willing to drill an extra whole into the firewall. This comes with security implications and I would not recommend this approach.
Hi Serhan,
Thanks for asking! To integrate subsystems with SAP EWM there has always been the Material Flow Interface (MFS). This is a real-time interface good for integrating any subsystems such as conveyor belts, fork lifts, etc. You can integrate robots, too. Unless the vendor of the subsystem provides an out-of-the-box integration, integrating is a project of weeks or months.
With Cloud Robotics we take a different approach for collaborative robots. We take advantage of the ROS-stack (based on Linux) most of these robots support. Using Google's open source tool Kubernetes, we treat the robot as an edge device that can take docker containers. These containers can be independent from a specific robot and the contain the EWM task. Thus, the warehouse order is copied to the robot and synchronizes with the robotics cloud using custom resources. Nice: The robot always knows what it does. Even if the Wifi connection breaks down for a few seconds or a minute, this does not pose a problem. The robot simply synchronizes when the connection is restored and meanwhile continues executing its task.
Yet, more important is the fact that thanks to the docker containers, this works with virtually any robot. Think of a demand peak in the warehouse such as Black Friday or Cyber Monday: You can add a robot within minutes and it simply integrates.
Kind regards,
Albrecht
Hi Albrecht,
That's actually helpful. Thanks for detailed explanation.
Best regards
Serhan
Albrecht
Cloud Robotics represents a departure and new alternative from the traditional approach of using MFS - in addition to COBOTS - can the Cloud Robotics be also considered as an alternative to communicate with - lets say an ASRF - conveyors ?
Regards
Mauricio
Hi Albrecht
Could you please provide any examples of customers out there that you may know using Cloud Robotics for EWM to integrate COBOTs?
Regards
Hi Mauricio,
we are just implementing at Bechtle Logistics and hope to go live this quarter.
Kind regards,
Albrecht
Albrecht
What it is the easiest way to find out a list of already supported OEMs?
Thanks
Mauricio
We already worked with MiR (Mobile Industrial Robots), Fetch, and Gebhardt.
Yet, if you see a robot vendor you would want to work with, we are happy to assist. With a relatively low effort on both sides we can make this working.
Hi Albrecht,
have you gone live with Cloud Robotics solution at Bechtle Logistics?
If yes, what are the experiences?
Regards,
Robert
Hi Robert,
Yes, Bechtle is live on Cloud Robotics. The key learning was that while the effort for robot integration has been minimized, the other work should not be overlooked. When an organization introduces robots the first time, there are some best practices that should be followed. These practices are not really new. Think of the robot as a new machine that is installed. Examples:
All these things have little to do with the EWM integration and are part of the job when installing new machines.
Dear Albrecht,
This is really very interesting topics and I have explore the GitHub but could not able to find the right documents to start.
Do you have how-to-guide to integrate EWM with Robots (may be LEGO MINDSTORMS)?
Best Regards,
Anil Kumar
Dear Anil,
did you take a look at the documentation on Github?
A very highlevel overview is also in part 2 and part 3 of this blog.
A Lego Mindstorm for deployment? I do not know whether this is impossible but for sure it won't be easy. The robot needs to have a Kubernetes Cluster.
For the MiR100 and the MiR250 (both by Mobile Industrial Robots), we installed an Intel NUC on the robot. We do not deploy the Kubernetes Cluster on the manufacturer's compute unit and therefore have no liability risks. Such liabilities could emerge if we would tamper with the original hardware. By using a separate compute unit we avoid this.
Of course, there are alternatives to the Intel NUC; such compute units are commodities.
By the way, stay tuned for a SaaS solution next year, which makes all of this even easier.
Kind regards,
Albrecht