1. Redefining robots. Too much focus on humanoid robotics is a misplaced definition of objectives. A robot is really a modular, autonomous platform combining measuring circuits, active RFID, RF, acoustic, laser and power technologies for in-situ data processing and real-time communication. Distributed architectures with this type of robot are revolutionizing industrial automation.
2. Don't fight the laws of physics. Adjust your expectations to what the robot is actually capable of doing.
3. Safety first. Pay attention to the cell layout from the operator's perspective. Fence the area so the operator cannot reach into the cell while the robot is in full motion. Make sure all cell doors have safety interlocks and light curtains as needed, and make sure you make the robot aware of these safety devices so it can do the job better. When it comes to designing your robot cell (electrical and mechanical layout) and programming, always consider safety as your top most priority. This aspect is even more important given all the codes and regulations for robotic safety. For more information on new safety requirements, visit http://awgo.to/026
4. Complexity can be barrier. When implementing robotics, engineers need to ensure the system is scalable, will mitigate safety and security risks and has built-in energy management. The robotic system must provide personnel at all levels with access to prognostic and diagnostic data in the same way discrete and process automation systems deliver operating data. Unfortunately, the complexity of the design process can often be a barrier to achieving the cost-savings and efficiency gains expected with robots.
5. Tooling clearance. It is very important to have a clear, defined understanding of the type of end effector for your application so that tooling does not damage in-order production parts. Selecting a well thought-out end effector will also help minimize the teach time on the final program to ensure the proper clearance to run the robot at the maximum speed.
6. Plan for maintenance. For large robots, make sure means for maintenance are included in the original project scope. A monorail hoist designed in at the start of the project, for example, is much cheaper and easier to install than one built years later. Installing a hoist while running requires scheduling, downtime, lost production, etc. Installing a hoist during emergency maintenance never happens, and temporary, sub-optimal, rigging ends up hurriedly being used.
7. Get early feedback. Try and do as detailed a simulation as early as you can and show it to the workforce involved in the process in order to get feedback on possible operational issues. You may find that there is more going on in the existing process than you knew about and that micro-management is necessary.
8. Zoning for product size. In a robotic pick-and-place system for thermoformed packaging trays, incorporate zoned suction cups that can be isolated or turned on and off, to allow for various tray sizes.
9. Electronic changeovers. Customization is only a fraction of the normal line production, but the ability to create electronic changeovers between formats in a small footprint streamlines the customization process.
10. Error codes. When you are programming a robot, many lines of code are involved. If you build in your own error codes in the program, it is easier to identify where in the program the error has occurred and to diagnose the issue. If it is a new issue, add it to the list of your error programs.
11. Faster re-mastering. For a robotic cell, design in a location datum point into the cell. Having an independent datum point outside of the usual work area (still within the robot work envelop) will allow quick re-mastering of an axis or robot following maintenance. This is critical for users of just a few robots, who don't typically have a lot of robot axis mastering experience. Witness marks on joints can fall off, and having an independent means to establish the coordinate system is a major time saver.
Tips for robotics integration.
Whether you're looking to integrate robotics into existing equipment or scale your control and information platform to accommodate robots, here are some general tips on the three primary options:
1. Use a single control platform that can be scaled to fit a wide range of robotics applications, regardless of size or complexity. This method allows the highest level of integration because it combines kinematic robot control within a machine's controller. All configuration, programming, kinematics, troubleshooting and operations are performed within a single control platform, which helps reduce engineering costs, training, maintenance and the overall machine footprint.
2. Use a single network technology and a common control and visualization environment. A networked approach integrates the robot control system with the machine control system. This is the most cost-effective solution for quickly integrating robotics into an existing application. Doing so gives the machine's controller access to the robot's control system, including diagnostics, necessary automation interlocks, troubleshooting, alarming and reporting.
3. Use a common control engine and development environment to help eliminate the need for separate controllers and systems. This embedded approach to robotics integration brings the robot module directly into the control platform's chassis. It keeps machine and robot control separate, but helps drastically reduce the machine footprint - by up to 50 percent - because there are fewer control boxes on a machine.
Mechatronics for OEMs
The perception of robots in the industrial workplace follows a form factor that leads users towards a limited amount of mechanical models. This feels modular, but it's not mechatronic. Changing the definition of robotics to coordinated axes synchronized by a single software and processor opens up conventional machine design to modular code and mechanisms--or mechatronics. OEMs that have adopted this new outlook make robotic machines that are faster, smoother and more flexible, not by changing their mechanics but by changing their approach. The choice for OEMs is even easier because they need only to migrate their control philosophy to those capable of true mechatronics.
Liked this article? Download the entire playbook here