Asked

Any Chance We Can Incorporate Moasure To Work With ARC

Any chance we can incorporate Moasure into our robots? This would be a big deal in many ways with its spatial awareness. If there was a way it incorporate this with ARC and an end effector it would solve lots of things.


ARC Pro

Upgrade to ARC Pro

Unlock the true power of automation and robotics by becoming a proud subscriber of Synthiam ARC Pro.

PRO
Synthiam
#1   — Edited

ARC is a framework, and we do not add products to the framework. Product support is added as robot skills, which are 3rd party and "can" be maintained by synthiam for a cost to the manufacturer. Companies add Synthiam ARC support to their product by creating robot skills or hiring us to create robot skills. Because that's how they get users of their product, you can email them and ask for an API and see if they even have the ability for 3rd party integration; I have no idea what the product is you're asking about, so I can't comment on what they do or have available.

Read more about how robot skills are integrated into ARC by 3rd parties: synthiam.com/Partner

#2  

Further to DJ's response, you have mentioned "end effector," which is a bit out of context. But, you may be referring to inverse or forward kinematics which calculates an end effector cartesian coordinate. You can do this with an existing robot skill here: synthiam.com/Support/Skills/Servo/Inverse-Kinematic-Arm?id=21839

Additional hardware is not required for calculating inverse kinematics because the physical robot dimensions are known and do not change. And for forward kinematics, the joint angle degrees are known.

#3  

Take a look at what it can do https://www.moasure.com

User-inserted image

Yes I understand Synthiam's position on this but it would be very cool to have these two companies work together.

PRO
Synthiam
#4   — Edited

I'd imagine that the company's agenda is to sell a product that connects to an iPhone for landscaping. Also, their website specifies how the system works and will not provide real-time data.

API Link here: docs.moasure.com/en/api/app-links-api

Quote:

Moasure has an App Link / Deep Link API that allows partners to build deep link functionality. This allows partners to, for example, create a button in their app to start a measurement, which, on tap, will open the Moasure app and start a measurement immediately. Once the action is completed, the Moasure app will redirect the user to the partner app with the relevant data that the partner app can parse. There are also other operations supported.

As synthiam support said, something like that has no usefullness on an end effector. Specifically, because it'll always have a growing error rate, as seen in their website screenshots. The error rate will continue to grow the longer the system is measuring. You can learn a lot more about how these types of systems work by reading this support document and further reading links provided within it: synthiam.com/Support/ARC-Overview/robot-navigation-messaging-system

What the error rate is from is called dead-reckoning; read more: www.oxts.com/dead-reckoning/

I covered this in detail in the past with other conversations, so I don't have much time to repeat myself in its entirety. The Coles notes version is that with that kind of measurement system, the error continues to grow the longer it is measured. In the use-case of that product, the operation time is minutes, not days or hours, and their website has a screenshot showing the error rate. This is why robotics never uses similar systems to calculate end-effector cartesian positions.

This is because the cartesian coordinate of the end effector can easily be calculated with forward kinematics -OR- the joint angles can be computed to reach a coordinate with inverse kinematics. This was explained above by Synthiam support, and they provided a link. With math calculations, the end effector will ALWAYS be 100% accurate and can never fail.

User-inserted image

  • 0.6% of 528 ft is 3.17ft
  • 0.6% of 12611ft2 is 75.67ft2

That is poor, and the error will only increase over time and from complicated movements. I think the product is good if you're using it to estimate a yard for planning a fence. I wouldn't trust it for moving a robot arm that could knock someone out or put a hole in my wall:)

Either way, I don't see their product as anything interesting. Perhaps you'd be interested in something like the T265: synthiam.com/Support/Skills/Navigation/Intel-Realsense-T265?id=20067

I might be totally off base by not fully understanding the usefulness of that product for robotics. You didn't provide any examples of what you'd do with it or what problems it would solve. So, I might be off base with my speculation.

#5  

I did not realize it was that far off, sorry wouldn't have bothered you with it. A different product that I was looking at was within cm level precision but only outdoors.

PRO
Synthiam
#6  

I'm still unsure what you would do with something like this. Do you have an idea of what you'd use it for? What problem does it solve etc.... That's more important to me because it helps me understand if there's something I can suggest. Or are you looking for toys to play with?:)

#7  

I envisioned using it on the driveway for various items, blowing the leaves and pine straw off, power washing the driveway, going to the end of driveway with large garbage barrel to put it there and retrieve it. Using it on the lawn to mow it, cut the edges, various other things. A lot of these things can be done with existing technologies but I did not have good results with them so far. A cm accurate GPS with RTK would solve a lot of these scenarios.