Roles No Roles
Full Name Etienne Samson
Job Title Director,Technical Support
Country CA - Canada
Company (optional) Robotiq
@mhowe add a counter variable to your program, put an if / else logic, the camera locate is in the if, the counter is in the else. Thus you will count how many time you do not see a part. You could use this so that after 5 or 10 photos without objects you can trigger your part feeding system.
Here are a couple ideas about time (and money) waste that are far from necessary in my opinion.
Some example of improper thinking during the design or early integration phase:
1) Spending time to adapt to inadequate equipment. I have often seen integration project where you need to adapt to an old, obsolete part of machinery that is not really meant to fit with your new equipment, you want to save money by re-using that old equipment but it is so complex to adapt to it that you spend 2 months "patching" the integration with software and jigs. That's a good waste of time and money were in a lot of case it would be a saver to just discard this and start with new machinery.
2) Same kind of situation where I've seen that the new equipment was not chosen properly for the application and then you fight to make it work for weeks. This happens especially when the management / buyers have not validated the choice with the engineer in charge of actually installing the equipment. Instead of accepting that there was a mistake in buying that equipment, people will work really hard on work-around to make it work where they would save by just returning the equipment and going back to the design phase.
3) Not using the "human experience" behind the process to automate. Most of integration process will automate something already done by a human operator. It is often very good to spend some time watching the operator doing his job, then translate this to an automated process. They often have tricks that are not written in the job description that can save an enormous amount of time!
Here are some examples of bad integration practices that can cause a lot of waste:
1) Not having a software versioning system. Getting lost in your program versions (when you reach rev_29 and so on) and loosing time trying to find the latest of your program version that all have the same name...
2) Not having a schematic view of your process. You need to keep your process close to what bring values, moving your part around without purpose close to the process can cause a lot of waste. Most integration projects can fit on a couple of post-it in-front of your desk. Part presentation, a couple of process related step, then part output. Any step in the program not linked to that is just waste.
3) Lack of internal communication: As soon as you have more then 1 working on the project, you need everyone on the same page. It's quite good to have a board where you can write down the automation project in a few blocks and have everyone work towards the same goal. Make sure that the mechanical guy working on the part presentation jig knows what the guy programming the robot needs for adequate part presentation etc. Working like software engineers with scrum meetings is probably the best thing to do.
@cdoeksen interesting topic you got there, I vote yes, those application do generate a lot of request and they all represent challenges with big potential. I'm not sure if each one could be as big as welding, but the sum of all them definitively can in my opinion!
@Sebastien @Student with the Robotiq wrist camera you should just teach the object on a simple uniform background. When in production you can look at the belt and if the light conditions are good you should be able to set a high detection threshold so that the part is recognized and the background is ignored properly.
Teaching the object on the textured background is probably not a good idea.