Advancing technologies have given robots the ability to adapt to the environment around them, greatly increasing their value in production applications in manufacturing.
Watching a robot in action is a pretty cool thing. A six-axis robot can perform moves that are as good, or better, than the human arm, and it can do it fast! We regularly wow students on facility tours who get to see our robot demos in action, zipping from one position to another. I imagine this is the case for automation professionals as well. When touring a manufacturing facility and catch a glimpse of a robot in operation, I eagerly await the opportunity to stand in front of it and watch it do its thing. Maybe this is just my reaction. Am I the only robo-geek out there or do you feel the same?
Unfortunately, more often than not, I am let down by what the robot is actually doing: the same thing over and over, following the same path at the same speed. Beyond the initial wow factor generated by a lot of motion and maybe an innovative end-effector (end-of-arm tool), robots are not really doing anything very cool or providing the value they could be if they were to leverage advancing technology.
Machine vision/3D laser scanning and tactile/force control are two technologies I am excited about.
I am excited that many integrators, original equipment manufacturers (OEMs) and robot manufacturers are embracing advanced technologies, and we are seeing more and more advanced robot applications on the manufacturing floor. When I go to trade shows these days, it is not uncommon to see robot demos that have the ability to adapt to the environment around them. In fact it is uncommon anymore to see a robot demo that does simple pick and place, or fixed speed/path operations.
Of course, there is a pretty big gap between the trade show floor and the production floor. It is one thing to show off cool features of a robot in a tightly controlled environment with the programmer sitting 10 feet away and another to have that same technology run reliably in a steel foundry with heat, dust, water, slag and other variables. That said, the reliability of some of these technologies is at a point where the technologies are indeed ready for primetime production applications. We are successfully deploying them now, as are others. It is time to take a serious look at the potential gains being offered.
Machine vision/3D laser scanning and tactile/force control are two technologies I am excited about. By adding “senses” to the robot, it opens up a whole new world of value that the robot can provide to manufacturing processes. Imagine the possibilities:
Bin picking and flexible infeed systems – Traditional infeed systems require parts to be singulated and precisely positioned in a fixture for the robot to act on it. The systems are mechanical in nature and require contact with the part. As such, they are significant maintenance points, a source part quality issues, and typically labor intensive when they are adjusted for different parts, driving up change-over downtime. By adding 3D laser scanning to the robot, parts can be left in the bin or on a pallet, where they are identified and located, so that a robot can go and grab them. These applications are called bin picking and depalletizing, respectively.
Another potential solution, where parts do not need to be singulated or fixtured, is to simply convey parts from upstream processes to a flat belt, where they can be scanned and picked as they are conveyed. There also exists the possibility of adding more flexibility to the overall manufacturing process, by eliminating fixturing and adding the ability to identify/locate parts, manufacturers can convey random parts and count on the robot system to sort it all out. These types of solutions result in significant maintenance, quality and downtime savings.
Adaptive control – Welding, friction stir welding and machining are all being done with robots these days. With tactile feedback, or force control, the robot can precisely control, and measure, the force that is applied by the robot to the part being worked on. So, for example, as the welding rod is consumed, the robot automatically compensates so the rod stays in contact with the part. Not only does this ensure the weld is completed, but it also results in a better quality weld by ensuring more consistent force at the weld point. In a similar fashion, friction stir welding is being done with robots by controlling the force of the tool through the parts as they are being welded.
While friction stir welding with computer numerical control (CNC) is not new, by using a robot, more complex shapes can be welded due to the force control along all the various vectors encountered. For machining, force control is used to adjust the speed through the part, slowing the process down as the tool becomes worn, reducing tool breakage and wear, while maintaining a consistent machining quality. These systems can even be programmed to notify the operator of the wear so that a tool change out can be scheduled. Adaptive control offers results that are measured by reduced downtime, improved quality, and lower maintenance costs.
Safety – Some robots are being delivered out of the box with force control sensors, such that they can be programmed to shut down if they are bumped. While this is a neat feature, I would not recommend it for any heavy lifting or high speed applications, as it is my philosophy to prevent the bump in the first place. That said, this type of feature does represent a nice secondary safeguard for personnel and equipment. For primary protection, laser scanning is a viable option that can be used to detect objects within the operational zone and either moving around it (collision avoidance) or shutting down. With these types of systems, traditional guarding and barriers can be eliminated, opening up areas for operators and maintenance to access equipment safely. Safety systems offer obvious operator improvements, but also can represent significant maintenance benefits.
These two technologies — machine vision/3D laser scanning and tactile/force control — integrated with robot workcells demonstrate that there are many opportunities for operational improvement that can directly impact a manufacturer’s bottom line through improved efficiency, reduced waste, improved safety and lower maintenance.