Automation (think timers & robots) + Robotics (designing robots) = More control over tasks!

 

Hi fre,
Do you have a question?

Moderation

 

Now that "Frelan has got your attention, I have been wondering about how somebody sitting in an office can accurately program a robot to work on a production floor. I have programmed robots to load parts into a fixture, and programmed them to perform tests that had required a human to work a lever and push a button. But if one does not have the locations specified to within 0.01mm, how does one specify the six positions of those non-orthagonal axis??I did it by using the control pendant to move mister robot towards the correct positioning , saving a few points as he got closer to the position.
Or are the programs created of a different sort, independent from fixed positions. That would have been the "Ro-Butt" seating testing robot that did endurance tests, all of the different ones, in the quality contro/source approval part of seating testing. AllI did was fix those robots when others broke them.

 

Industrial Robot applications are primarily limited by the robot positioning accuracy and motion range, and the required production rate. Surgical robots are limited by size and feedback, and then by accuracy. Other robotic applications are mostly limited by cost effectiveness.

And in every case robotic applications are limited by what resources the robot has available. That exposes the fallacy that automation is the solution to everything.

 

Now that "Frelan has got your attention, I have been wondering about how somebody sitting in an office can accurately program a robot to work on a production floor. I have programmed robots to load parts into a fixture, and programmed them to perform tests that had required a human to work a lever and push a button. But if one does not have the locations specified to within 0.01mm, how does one specify the six positions of those non-orthagonal axis??I did it by using the control pendant to move mister robot towards the correct positioning , saving a few points as he got closer to the position.
Or are the programs created of a different sort, independent from fixed positions. That would have been the "Ro-Butt" seating testing robot that did endurance tests, all of the different ones, in the quality contro/source approval part of seating testing. AllI did was fix those robots when others broke them.

Perhaps this is where you may be outdated engineering wise. I think it is very much the case where modern intelligent programs can be written without having to physically interface with the work piece(s). Lidar is a neat example which can theoretically image a surface much better than a system relying on moving parts. If I'm able to build such a sensor and then build machines which allow me to control it remotely, how is this less sophisticated than a classical engineering approach like you described?

 

A surgical robot is a good example of where the positioning isn't known until the last minute. Obviously 3D CAT/MRI scans give lots of detail in planning the op, but at the end of the day the surgeon makes the decision based on what he sees and the haptic feedback of what he feels whether local or remote.