Menu

Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - Johannes @ Robots in Architecture

#721
Support / Re: Restrict A1 movement
October 03, 2018, 11:17:48 AM
Not with a six-axis robot as you do not have the required redundancy.

If you put the robot onto a rail (= 7 axis system), you could e.g. move the rail in a way that there is minimal A1 movement.
With a 6 axis robot you can just lay out the frames in an efficient way that improves reachability and reduces A1 movement. But there is no standard strategy or pattern you could use - that very much depends on what you are using the robot for.

Best,
Johannes
#722
Tutorials / Re: Surface Milling Tutorials
October 01, 2018, 11:59:42 AM
Hello,

I haven't come across any GH plugin that would do that for you in a reliable way, as those strategies are quite complicated and also do not fit that well into the GH way of programming.
Don't get me wrong, it's definitely possible to create milling strategy for individual use cases in GH and you can save great amounts of time through mass customization opposed to CAM software. But there isn't a script that will work well for "all" geometries, like in CAM software where you define the stock, the geometry, set a few parameters and get a working solution.

For "generic" CAM work we usually use Fusion 360, as it has got a nice interface and doesn't cost much (free for EDU). There is a Fusion 360 component in KUKA|prc, you can right-click it to export a postprocessor. Code generated through this PP can then be imported into GH and e.g. combined with your parametric geometry. Note that you need an additional "orientation" point as you get 5-axis G-code out of Fusion, that does not define the rotation around the tool axis!

Best,
Johannes
#723
Support / Re: Sanding elipses using a fixed tool
September 28, 2018, 05:12:54 PM
Hello,

Here's the file in Rhino 5 format. The GH file will complain about a version mismatch but should work as intended!
I actually didn't make it to China, here's a video what the colleagues from IP did: https://vimeo.com/287454509

Best,
Johannes
#724
Support / Re: Sanding elipses using a fixed tool
September 24, 2018, 06:40:25 PM
Hello,

Here's a quick example that should help you get started polishing the wood ellipses.
You see that A6 is running into its limitation, so you may need to set A6 on your robot to infinite rotation, which of course complicated the air supply for the gripper.
Maybe make sure that the robot starts close to the negative limit and then only turns in the positive direction.

Best,
Johannes
#725
All the best for the future, Akanksha!
#726
Here's an example to illustrate it.
The first PTP and LIN positions are outside of the A5 reach, so that A5 is 1. The next position cannot be reached at all, which is why all axes are at 1. The axis value that you are getting is only the (more or less) maximum extension so that you can see where the unreachable position is located at. I believe that's better than just having the robot disappear when there is no mathematical solution.

Best,
Johannes
#727
Hmmm... I wasn't able to replicate that, see the attached screenshot.
It may be that your position is completely out of range, i.e. unreachable, no matter the axis limit? Like e.g. when you put a position 10.000mm away from the robot's center.

Best,
Johannes
#728
Hello,

The "KUKA KR150-240 -2" should be the right one. But there are also additional versions like e.g. KR150L130 that are longer and therefore support less weight.
To be on the safe side, move your robot into its standard home position with A2 at -90 and A3 at 90 and then get its Cartesian position with both tool and base set to 0.
Make a point with those coordinates in Rhino/GH and check if it corresponds to the digital robot.

Best,
Johannes
#729
I'll look into the all 0/1, but the ten values refer to a maximum of 10 axes, 6 + 4 external ones.

Best,
Johannes
#730
Hello,

For me the problem with NaN or null is that it doesn't tell you if it's exceeding the range in positive or negative direction.
We already provide the Reachability output in the Analysis component - doesn't that do the job?
You could just replace the axis values with NaN based on that output, if you want to.

Best,
Johannes
#731
Hello,

Well, I'd say it depends on which direction you want to take, if you want to work at a company, found a startup or do academic research.
I'm guessing that you are outside the EU and US, so it may make sense to enroll for a relevant Master or PhD program, also to acquire skills with robots. There are several countries in Europe with good academic programs and very low or free tuition. Note that there are usually restrictions on how much you are allowed to earn with a student visa.
I don't think your nationality is a problem - you just need to be allowed to work in the EU or UK, you don't need to be from there. I would assume that if you finish an academic program it would be easier to get a work permit in that country, but I'm really not an expert.

Hope that helps a bit!
Best,
Johannes
#732
Sounds good, will look into that. I'll need to look into the most understandable CNC terms for it. Probably Rapid for Flyover, Approach and Retract?
Best,
Johannes
#733
Support / Re: Collision and Unreachable Position Detection
September 13, 2018, 12:00:55 PM
Hello,

Can you maybe internalize the plane input? That way it's static geometry and we don't need the custom DLLs.

Best,
Johannes
#734
Support / Re: Collision and Unreachable Position Detection
September 11, 2018, 04:27:58 PM
Hello,

My guess would be that the unreachable collision occurs in-between two movements. If you disabled it, make sure to enable the Interpolate Toolpath option in the Advanced Settings, and if it's on, try changing it to a lower value like 1mm.

Also, you get the collision value for each joint, so consider using Mass Addition to sum them up!

Hope that helps!
Best,
Johannes
#735
Support / Re: LIN Speed Synchronization including PTP
September 03, 2018, 06:22:11 PM
Hello Sam,

I don't quite understand the idea, you want to have LIN and PTP motions with the same speed? That will be tricky, as they probably also have different acceleration and deceleration curves.
PTP is always set by the slowest axis involved in a motion, so movements can have very different speeds.
If you describe what you want to achieve, maybe I can help you find a walkaround.

Best,
Johannes