Team Delft’s Robot Winner of the Amazon Picking Challenge 2016 Carlos Hernandez 1(B ) , Mukunda Bharatheesha 1 , Wilson Ko 2 , Hans Gaiser 2 , Jethro Tan 1 , Kanter van Deurzen 2 , Maarten de Vries 2 , Bas Van Mil 2 , Jeff van Egmond 1 , Ruben Burger 1 , Mihai Morariu 2 , Jihong Ju 1 , Xander Gerrmann 1 , Ronald Ensing 2 , Jan Van Frankenhuyzen 1 , and Martijn Wisse 1,2 1 Robotics Institute, Delft University of Technology, Mekelweg 2, 2628 Delft, CD, The Netherlands c.h.corbato@tudelft.nl 2 Delft Robotics, B.V., Mijnbouwstraat 120, 2628 Delft, RX, The Netherlands Abstract. This paper describes Team Delft’s robot, which won the Amazon Picking Challenge 2016, including both the Picking and the Stowing competitions. The goal of the challenge is to automate pick and place operations in unstructured environments, specifically the shelves in an Amazon warehouse. Team Delft’s robot is based on an industrial robot arm, 3D cameras and a customized gripper. The robot’s software uses ROS to integrate off-the-shelf components and modules developed specifically for the competition, implementing Deep Learning and other AI techniques for object recognition and pose estimation, grasp planning and motion planning. This paper describes the main components in the system, and discusses its performance and results at the Amazon Picking Challenge 2016 finals. Keywords: Robotic system · Warehouse automation · Motion planning · Grasping · Deep learning 1 Introduction The Amazon Picking Challenge (APC) was launched by Amazon Robotics in 2015 [3] to promote research into robotic manipulation for picking and stocking of products. These tasks are representative of the current challenges that warehouse automation faces nowadays. The unstructured environment and the diversity of products require new robotic solutions. Smart mechanical designs and advanced artificial intelligence techniques need to be combined to address the challenges in object recognition, grasping, dexterous manipulation or motion planning. Amazon chose 16 teams from all over the world to participate in the finals at RoboCup 2016. Team Delft won both the picking and the stowing challenges. Section 2 discusses Team Delft’s approach, explaining its design principles and c  Springer International Publishing AG 2017 S. Behnke et al. (Eds.): RoboCup 2016, LNAI 9776, pp. 613–624, 2017. https://doi.org/10.1007/978-3-319-68792-6_51