No More Heavy Lifting: Robotic Solutions to the Container Unloading Problem

Abstract

This article discusses the scientifically and industrially important problem of automating the process of unloading goods from standard shipping containers. We outline some of the challenges barring further adoption of robotic solutions to this problem: ranging from handling a vast variety of shapes, sizes, weights, appearance and packing arrangement of the goods, through hard demands on unloading speed and reliability, to ensuring fragile goods are not damaged. We propose a modular and reconfigurable software framework in an attempt at efficiently addressing some of these challenges. We outline the general framework design, as well as the basic functionality of the core modules developed and present two instantiations of the software system on two different fully integrated demonstrators. While one is coping with an industrial scenario, namely the automated unloading of coffee sacks, with an already economically interesting performance, the other scenario is used to demonstrate the capabilities of our scientific and technological developments in the context of medium-to long-term prospects of automation in logistics. We performed evaluations which allow us to summarize several important lessons learned and to identify future directions of research on autonomous robots for handling of goods in logistics applications.

Self-supervised damage-avoiding manipulation strategy optimization via mental simulation

“…The possibilities are manifold: Firstly, the constraints on the motion planning search space are relaxed in comparison to approaches where other objects are considered as obstacles. As in previous work [1] which uses a conventional motion planning method without sophisticated task-level planning, some scene configurations do not allow for collision-free manipulation at all. The presented method, however, works one level higher than motion planning and allows for generating manipulation strategies which facilitate the motion planning itself.…”

Visual Object Categorization Based on Hierarchical Shape Motifs Learned From Noisy Point Cloud Decompositions

“…It provides a valuable feedback about the actual generalization capability of the proposed approach which is a crucial property for applications in real world scenarios. For instance, in the field of robotic logistics [23,52] instance appearances may encounter strong variations (as they were drawn from different distributions) due to unstructured and confined spaces, clutter or even limited maneuverability or kinematic constraints of the robot platform, see Fig. 21.…”

Grasp quality evaluation done right: How assumed contact force bounds affect Wrench-based quality metrics

“…The hand's 7 DoF are controlled via 7 motors that allow independent joint torque control. metrics used in grasp planners [3], [4], [5], in synthesis algorithms based on prototype grasps [6], [7], [8] as well as in grasp success prediction systems [9].…”