Robotic arm demonstrates depalletising of different sized boxes

Californian startup, Kinema Systems has demonstrated its Kinema Pick robotic arm. The company claims it is “the world’s first self-training, self-calibrating software solution for robotic depalletising”.

Depalletising is a simple process to automate with an industrial robot arm so long as each pallet is the same size and contains the same boxes in the same order. The Kinema Pick robotic arm can handle the picking of unstructured random pallets of boxes by combining 3D sensing with advanced motion planning.

What’s more, the robot requires no further training past setting up the vision system, telling it where the pallet is going to be and where it needs to move the boxes. Each time the arm picks up a box it creates a model of what the box looks like, eliminating manual training time, minimising integration time and helping it to pick quicker over time.

Kinema Systems was founded by Sachin Chitta and Dave Hershberger, both of whom worked at Willow Garage and SRI and helped create open-source software for robotics including MoveIt!, ROS-Control and Rviz, used by hundreds of companies, researchers and universities around the world on robots ranging from industrial automation systems to humanoids.

The company’s long-term vision is to build easy to use software solutions for advanced robotic manipulation applications in industry and beyond. It has started pilot testing with a number of large material handling companies. Once these tests are complete, the Kinema Pick is expected to be made available by the end of 2016.