Flowstate: Intrinsic application to simplify the creation of robotics applications

Finally, Essential (a spin-off of Google-X) have revealed a product they’ve been working on with help from Open Source Robotics Company team (among others): flow conditions!

What is Flowstate?

Introducing Intrinsic Flowstate | intrinsic (image copyright by Intrinsic)

Flowstate is web-based software designed to simplify the creation of software applications for industrial robots. The application provides an easy-to-use desktop environment in which blocks can be combined to define the desired industrial robot behavior for a particular task.

Good point

• Flowstate offers a variety of features, including simulation testing, debug toolsand seamless deployment to real robots.
• He based on ROso we should be able to use our favorite framework and all existing software to program it, including the Gazebo simulation.
• It has a behavior tree-based system for graphically controlling program flow, which simplifies how to write programs only blocks move around. But it’s also possible to switch to expert mode to touch up the code manually.
• It has an existing library robot model and hardware ready to be added, but you can also add your own.
• In addition, the application provides pre-built AI skills which can be used as modules to achieve complex AI results without the need for manual coding.
• One limitation (but I do consider it a good thing) is that the tool is a thought for industrial robots not for service robots in general. This is great because it puts focus on the product, especially for this early release

Flow conditions | intrinsic (image copyright by Intrinsic)

Based on the official post and keynote released on Monday, May 15, 2023 (available Here), this is the information we have collected so far. However, we do not currently have a comprehensive understanding of how the software works, its full feature set, and any potential limitations. For more insights we’ll have to wait until July this year, hoping I’m among the lucky participants selected for the private beta (open call to beta still available) Here).

Point not clear

Even if I find the Intrinsic proposal interesting, I’ve identified three potential problems with it:

1. Interoperability across multiple hardware and software platforms poses challenges. Full OSRC team recruitment by Intrinsic emerged to address this issue, given that ROS is currently the closest system on the market to achieving such interoperability. However, widespread adoption of ROS by industrial robot manufacturers is still limited, with only a few companies implementing it.

   Ensuring hardware interoperability requires the adoption of a common framework by robot manufacturers, which is currently far from a reality. What we ROS developers are aiming for right now is for someone to be able to build a ROS driver for the robotic arm we want to use (such as for example a robot manufacturer, or a team from ROS Industry). However, manufacturers are generally hesitant to develop ROS drivers due to limited business potential and their aim to lock in customers. Unless platforms dedicate substantial resources to developing and maintaining drivers for supported robots, hardware interoperability challenges cannot be solved by platforms alone (in fact, that’s one of the goals ROS-Industry aims to achieve).

   Google has the potential to unite hardware companies towards this goal, as Wendy Tan White, CEO of Intrinsic, put it, “This is an ecosystem effort” However, it is very important for the industry community to experience real benefits and value in supporting this initiative beyond helping others in building their business. The specific benefits that ecosystems can gain by supporting these initiatives are unclear.



Flow conditions | intrinsic (image copyright by Intrinsic)

2. Availability of ready-made AI skills for robots is a complex task. Consider skills widely used in ROS, such as navigation or arm path planning, exemplified by Nav2 And Move, which offers great functionality. However, integrating these skills into a new robot is not as simple as plug-and-play. in fact, special courses exist to teach users how to effectively utilize the various navigation components within the robot. This highlights the challenges associated with applying such skills to robots in general. Thus, it makes sense to anticipate similar difficulties in developing pre-made skills in Flowstate.

3. The last point that I didn’t see clearly (because it wasn’t discussed in the presentation) is how the company will do business with Flowstate. This is a very important point for every robotics developer because we don’t want to be locked into a proprietary system. We understand that a company should have a business, but we want to be clear about what that business is so we can decide whether or not it’s convenient for us, both in the short term and the long term. For example, Amazon’s Robomaker hasn’t gained much traction for forcing developers to pay for the cloud while running Robomaker, when they can do the same (with less fancy stuff) on their own local machines for free.

Conclusion

Overall, although Flowstate shows promising results, more information and hands-on experience are needed to assess its effectiveness and overcome potential challenges.

I have signed up for the limited beta. I look forward to being selected so I can have first hand experience and report on it.

Be sure to read the original post by Wendy Tan White and the keynote presentation, both of which can be found at intrinsic network.

Flow conditions | intrinsic (image copyright by Intrinsic)