It's a set of two different-sized 6-axis arms on a low-profile base that can move about with Omni wheels

The central application is small tasks in a light industrial environment.

The designs here are all very much conceptual work. There are issues with them. See if you can spot the biggest one before scrolling down.

Working title: “Leon”

“Leon” is the name for this concept work.

Named after none other than the legendary:

+4 mins reading, +5 images

Leon the Lobster.

https://www.youtube.com/watch?v=UFOqxgYI8VM&ab_channel=BradyBrandwood

Lobsters being one of the few animals with bilaterally asymmetrical appendages…. They have a different specialized claw on each side.
But also because Brady Brandwood’s videos on Leon, are an inspiration. It's a reminder to try make something nice. Doesn't need to be dramatic or spectacular.

Scribble

Here is some early sketch work on my remarkable. https://remarkable.com/

Some functions

There are a few applications scattered around the sketch. Such as:

Pulling around a trolley of parts and extra tools
Mount points for expansion of parts,
This includes the idea that it can remove/replace its own batteries and parts to get lighter and increase its agility.
A flat design to fit under cars
Some ways to lift itself up to a table
A standard wall plug. The idea being, this robot is a mobile arm, but not really for all-day warehouse logistics. Its battery life should be enough to get it to its next workstation, where it would then plug itself in.

What is it good for?

It’s a helper bot.

The main purpose is to be a tool that is less work to configure than it is to do the same job yourself.

Tasks like “hold this part and screw these bolts into it”. “Go and organise the items on that table.” Etc.

The “what why and how” of this is to be discussed in future posts. (And linked here).

Its main areas of application are offices and light industrial environments.

It might still have values in the following areas. But these are not what it's to be a dedicated design for:

a high-performance machine for constant industrial use.
a toy
an educational tool
a research robot
for outdoor use like rugged terrains
To be better or faster than a human is at anything.

If you want to go faster and better. Your local cobot supplier is the way to go. I recently talked with a couple of them at a Universal Robots convention. More on that later.

Quick CAD work

Here is the first stage mockup on Fusion 360.

Here's a link to the cad software for personal use https://www.autodesk.com/products/fusion-360/personal

This project is not going to have revenue for some time. But I'll be buying this the moment something changes.

This was about 3 hours of work. CAD is not my strong point (by degree and career, I'm an electrical engineer) so I had to learn Fusion 360 from scratch.

This is a great tutorial: https://youtu.be/d3qGQ2utl2A

At the studio, we (someone else) use Solidworks for all our CAD needs. It's also great. But there's a bit of a trend of newbies to go with fusion, so I'm trying it.

How big is it?

While learning Fusion, I made a few pieces of common Ikea furniture in order to get some idea of scale, reachability and store-ability.

Here I determined a good “size” for an office robot if you want to store it easily when not using it.

Can this robot do the dishes?… no it probably cannot.

Can this robot be stored in most cupboards?… no it probably cannot.

Some improvements and features need to be fleshed out.

Issues:

Terrible range:

Don't get too creative and optimise too early.

In order to be “smart” with part reuse I flipped the 1st two axis. Another advantage is that it might make the total design pack-down flatter, giving advantages in transport and storage.

This is not how 6axis arms are typically designed. This is where being creative can be naïve.

The thought was “Why make the first joint on the z-axis, when you can just move around on the omni wheels?”

Yes.. This is now the rough range of dexterity.

Green = can grip in most angles in this area Yellow = can grab things, not only at limited angles Red = cannot reach at all.

Basically the working area is very limited and a lot of rolling around is needed to compensate for that.

Its even worse if you want to use two arms to do difficult work.

Any application on the platform itself gets greatly limited.

Finally, the classic, Axis 1 now needs to work harder because the weight of the motor on Axis 2 pulls on it.

Not a lot of room on the platform.

Clearly, the wheels are missing their motors, there are missing parts everywhere.

Robot arms are far from self-contained. The control unit, Motor driers, and power suppliers. It’s all big heavy stuff. So when you hear “128mm” footprint… that’s not the total footprint.

The UR3e, for example, has an external control box called (??? OEM?, e-series?.. control box?) that’s about 5kg and the footprint of a computer keyboard.

But those are probably great specs for the performance and reliability that they are known for… and high performance is not what this project is about.

What, is it just going to crawl around on the floor then?

yeah, I know right?… clearly some changes are needed.

To be continued….

Speaking of…

What’s with the dog reference?

A lot of dog robots getting made these days. … most of them just trotting around with cameras on them.

Perhaps a discussion for another time.