Vision Guided Robots for Fruit Picking: The Role of Embedded Vision

Vision guided robots (VGR) for fruit picking have yet to hit mainstream commercial applications but show strong potential as a near future solution. VGRs are essentially as the name implies – robotic arms that are guided by advanced vision systems.

This differs from typical industrial robots in that many robots are programmed to perform the same repetitive motions over and over. Industrial robots that do leverage vision systems are often meant to account for small variations in object orientation, or for things like seam tracking in robotic welding. VGRs actually guide a robotic arm as a central part of the application.

Technical Barriers to Robotic Fruit Picking

Fruit picking applications are challenging for robots. Not every piece of fruit looks exactly the same, some may be ripe and others may not be, and conditions outdoors vary on a daily basis. Essentially, imaging capabilities remain one of the last barriers to achieve robotic fruit pickings with VGRs.

Inefficient illumination, occlusion, or inaccurate distance estimation have been the primary constraints in terms of robotic fruit picking. Researchers are racing to find the right combination of 2D, 3D, NIR and hyperspectral vision solutions to be the first to achieve reliable fruit picking.

Certified System Integrator Program

Set Yourself at the Forefront of the Global Vision Market

Vision system integrators certified by A3 are acknowledged globally throughout the industry as an elite group of accomplished, highly skilled and trusted professionals. You’ll be able to leverage your certification to enhance your competitiveness and expand your opportunities.

GET CERTIFIED

Embedded Vision’s Role in Fruit Picking

Embedded vision is playing an important role in enabling fruit picking of all kinds, packing several imaging sensors onto one robotic system to allow advanced imaging for more adaptable robotic movement.

One promising combination includes 2D RGB vision and 3D vision technology to provide robots with important location and depth for guidance. With this information, the robot cannot only detect the presence and orientation of cherries but can determine the best location to shake a cherry tree to harvest the cherries.

Embedded vision is a vital component of this system. Without the ability to have multiple vision streams, as well as image processing capabilities, in a compact space there would be no fruit-picking VGRs.

VGRs have advanced dramatically since their early days in exclusively industrial settings. As new commercial applications sit just beyond the horizon, the pace of innovation is only accelerating.