Bin-picking Robot using a Multi-gripper Switching Strategy based on Object Sparseness

Bin picking of various daily items is an important research problem in robotics. If the target items are diverse, multiple grippers are normally used. A design of gripper combinations depends not only on the item variations but also on the state of the bins, which changes while robots pick items from them. In this paper, we propose a gripper combination strategy to change the gripper combination during a bin-picking task based on the sparseness of objects inside bins. As an experiment, we build a robot system which has three different types of grippers. By using the proposed combination strategy, the system effectively changed the gripper combination during the task, and picked 18/20 items to obtain the 3rd place in the Stow task at the Amazon Robotics Challenge 2017. The successful picking rate and Mean Picks Per Hour (MPPH) were higher than the 1st place team. In this paper, we describe the problem, method to switch the combination, system including gripper design and recognition algorithm, and experimental results from the competition.