The Future of Automated Random Bin Picking
Robotic bin picking is incredibly important as it is able to take the tedious, monotonous tasks off of workers hands. It is able to effectively handle bulk parts sorting, dangerous operations, and/or labor-intensive order fulfillment. Because of its complexity, random bin picking has yet to be perfected or completely grasped by automation systems. However, visionaries are pioneering their way into that uncharted territory. Many experts predict that robotic random bin picking will become mainstream by 2020 as there are a variety of subsets of bin picking are already commonplace.
A Staff Engineer at FANUC America Corporation in Michigan, David Dechow, says that the “Marketplace is hyped up about robotic bin picking. The reality is that machine vision is suitable, robust, and reliable for a subset of that entire world of things we would like it to do, whether it be inspection, 2D guidance, or 3D guidance. From my hands-on experience over a number of years, bin picking is in the same category now. There is real-world capability, but it is still a subset of the whole world of things we would love it to do.”
In general, bin picking can be divided into three different categories. Each contains a special amount of considerations based on the part position that results in is a different level of complexity, cost, and time. As one may guess, the characteristics of parts often determine how easily they will be picked from a bin. The parts that have the greatest success in Bin Picking are geometrically symmetric parts. These parts don’t have strange features, are not too heavy, and have some sort of sufficient planar surface in all of their random orientations. These few geometric surfaces are very easy to pick up with a robot and very easy to grip.
The three categories are as follows: Structured Bin Picking is when parts are situated in an organized pattern in which they can easily be identified and picked. This can be automated using 2D Vision, imaging, and analysis. Semi-structured Bin Picking is when the parts are positioned with mild amounts of organization to aid in picking. Random Bin Picking is when the parts have completely random positions, can be overlapping, and have multiple orientations, making this the most complicated version. Advanced technologies consisting of 3D imaging and 3D analysis will have to be created to tackle the most challenging parts yet; the shingled, packaged, or deformable parts that are difficult to image with machine vision. The height variation, thin nature, and particular geometry make it very confusing, even with 3D imaging techniques.
In order to continue to work towards success with 100% accuracy for random bin picking, there are three big components that need to work together. These are the sensors, software, and EOAT. Many companies are aware of this and are currently racing to create the best, most efficient software to provide the perfect random bin picking solution. They are in the works of ensuring the perfect algorithms to prevent things such as the robot and EOAT from colliding with the sides of the bin and other objects. FANUC has already stepped up to the plate and created an Interference Avoidance feature that is software based and comes standard with the 3D area sensor.
So far, there has been a successful turnkey (cell, robot, process automation equipment, vision, artificial intelligence, and safety equipment. etc.) solution for robotic pharmacy order fulfillment systems that can ensure 100% guarantee. The RX Unit Pick Workcell is a completely automated robotic pharmacy order fulfillment system that can handle thousands of specific, pre-packaged, medicine for mail order delivery or to fill for central locations. This workcell can make dispensing of pre-packaged, high volume units-of-use, look easy.
Another big area of growth for robotic bin picking is E-commerce and order fulfillment. Stores such as Walmart, Costco, or Target may eventually look to automation for their online ordering. The challenge right now is the end effector that needs to be able to grab a bottle of detergent and/or tube of lip balm. Soft Robotics is working on adaptive gripping technologies, seen with their flexible grippers combined with Universal Robotics Spatial Vision Software.
As you can see, it is so important for researchers to continue thinking, dreaming, and ultimately creating solutions to these everyday, random bin picking applications. At RobotWorx, we believe that the remaining applications that pose specific challenges will be solved in the near future. We are excited for what is to come! For any questions about automation or bin picking robots, contact us online or give us a call at 740-251-4321.