Grasping and picking of a wide range of objects is a canonical problem for robots. Industrial applications have been around for decades, yet robotic systems suffer from high failure rates in unstructured, changing environments. Warehouse automation provides a stepping stone from industrial (structured) to more unstructured pick and place scenarios. The Amazon Robotics Challenge (formerly Picking Challenge) has highlighted some of the issues since its inception in 2015. The challenges encompass various aspects of grasping, vision and other forms of sensing, gripper and robot design (including mobile manipulators), motion planning, optimization, machine learning, software engineering, and system integration, among others. The main aim of this workshop is to gather past and future participants of these challenges and the wider robotics and automation community to discuss their robotic solutions, experience of the previous competitions and their vision on automating item picking for warehouse logistics. In addition we would like to have the broader community taking part, such as researchers who did not (yet) take part in these challenges. Attendees of the workshop would have the opportunity to see the state-of-the-art in item picking research and development and interact with the people who are passionate about solving this complex problem during the presentations, poster discussions and open forums. This workshop is co-organized by Amazon Robotics, the organizer of the Amazon Robotics Challenge.