A critical part of implementing OLRP in a robot cell is the alignment of the virtual cell to the real-world cell. To dramatically oversimplify the process: you take points on your physical robot to map out key features of the cell, then move the virtual components to match those point locations. If we dig into this process a little deeper, the question comes up, what are these points relative to?
How does our virtual robot know where these points are?
There are two basic ways that robots define the location of a point in space:
The good news is either method will work, and both will produce essentially the same results. However, aligning using point coordinates offers some advantages for OLRP. Point coordinates advantages boil down to an overall better experience when using base frames for programming new jobs in the future.
In a previous blog post, Virtual to Reality: Delivering Accuracy to The Real World. We discussed the real-world factors that can negatively affect the quality of a robot cell alignment and some of the strategies that can be used to mitigate those factors. One of those strategies was to manually touch up points and programs which were not properly aligned to the workpiece. Having a program that is taught relative to a base frame (or base frames) can make this process quite a bit faster and easier. In the event the robot cell or workpiece is significantly misaligned, you could touch up a single base frame to adjust the location of an entire program of points at once to move them to the correct location, rather than adjusting them one-by-one. Having your robot cell aligned relative to a coordinate system and using this strategy can make a world of difference when it comes to your OLRP experience; especially in robot cells that are not equipped with other part-finding or vision systems.
In certain extreme cases, when workpiece CAD is significantly different from the geometry of the real-world parts, this strategy can be taken a step further by compartmentalizing robot points and paths within a program into different base frames. Similar to above, these frames can then be touched up to adjust the location of several points at once.
It is possible to align your robot cell based on joint values then still use base frames to create new jobs for your robot with OLRP. However, this creates more opportunities for errors to creep into the programming process.
The bottom line. It is best to remain consistent in how you define your points between alignment and programming to minimize errors. If you’re choosing only one system to define points, I recommend using coordinate systems and base frames.