Standing on Giant’s Shoulders: Newcomer’s Experience from the Amazon Robotics Challenge 2017

  title={Standing on Giant’s Shoulders: Newcomer’s Experience from the Amazon Robotics Challenge 2017},
  author={Gustavo Alfonso Garcia Ricardez and Lotfi El Hafi and Felix von Drigalski},
International competitions have fostered innovation in fields such as artificial intelligence, robotic manipulation, and computer vision, and incited teams to push the state of the art. In this chapter, we present the approach, design philosophy and development strategy that we followed during our participation in the Amazon Robotics Challenge 2017, a competition focused on warehouse automation. After introducing our solution, we detail the development of two of its key features: the suction… 
What are the important technologies for bin picking? Technology analysis of robots in competitions based on a set of performance metrics
This paper proposes a set of performance metrics selected in terms of actual field use as a solution to clarify the important technologies in bin picking and uses the selected metrics to compare the four original robot systems, which achieved the best performance in the Stow task of the Amazon Robotics Challenge 2017.
Bin-picking Robot using a Multi-gripper Switching Strategy based on Object Sparseness
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.
Restock and straightening system for retail automation using compliant and mobile manipulation
The proposed mobile manipulator features a custom-made end effector with compact and compliant design to safely and effectively manipulate products in retail stores to restock shelves, dispose expired products, and straighten products in Retail environments.
Packing Planning and Execution Considering Arrangement Rules
A packing robot that plans the arrangement and then executes it and uses the heuristics of the packing so that objects are stacked from the bottom and the strategy to approach from the top reduces the possibility of collision.


Lessons from the Amazon Picking Challenge: Four Aspects of Building Robotic Systems
This work describes the winning entry to the Amazon Picking Challenge, and suggests to characterize robotic system building along four key aspects, each of them spanning a spectrum of solutions—modularity vs. integration, generality vs. assumptions, computation vs. embodiment, and planning vs. feedback.
Analysis and Observations From the First Amazon Picking Challenge
An overview of the inaugural Amazon Picking Challenge is presented along with a summary of a survey conducted among the 26 participating teams, highlighting mechanism design, perception, and motion planning algorithms, as well as software engineering practices that were most successful in solving a simplified order fulfillment task.
Multi-view self-supervised deep learning for 6D pose estimation in the Amazon Picking Challenge
This paper proposes a self-supervised method to generate a large labeled dataset without tedious manual segmentation and demonstrates that the system can reliably estimate the 6D pose of objects under a variety of scenarios.
The SMACH High-Level Executive
This column introduces an approach based on nested state machines that has proven very effective at building real-ROS applications and explores the trade-offs between task scripting and task planning for highlevel control in robot applications written on top of ROS.
A Dataset for Improved RGBD-Based Object Detection and Pose Estimation for Warehouse Pick-and-Place
This letter provides a new rich dataset for advancing the state-of-the-art in RGBD-based 3D object pose estimation, which is focused on the challenges that arise when solving warehouse pick-and-place tasks.
Vibration-Reducing End Effector for Automation of Drilling Tasks in Aircraft Manufacturing
In this letter, we present an end effector that can drill holes compliant to aeronautic standards while mounted on a lightweight robot arm. There is an unmet demand for a robotic solution capable of
YOLO9000: Better, Faster, Stronger
YOLO9000, a state-of-the-art, real-time object detection system that can detect over 9000 object categories, is introduced and a method to jointly train on object detection and classification is proposed, both novel and drawn from prior work.
Towards MRI-Based Autonomous Robotic US Acquisitions: A First Feasibility Study
This work presents a set of methods and a workflow to enable autonomous MRI-guided ultrasound acquisitions and uses a structured-light 3D scanner for patient- to-robot and image-to-patient calibration, which in turn is used to plan 3D ultrasound trajectories.
You Only Look Once: Unified, Real-Time Object Detection
Compared to state-of-the-art detection systems, YOLO makes more localization errors but is less likely to predict false positives on background, and outperforms other detection methods, including DPM and R-CNN, when generalizing from natural images to other domains like artwork.
On the Performance of the Intel SR30 Depth Camera: Metrological and Critical Characterization
The main aim of this paper is to characterize and to provide metrological considerations on the Intel RealSense SR300 depth sensor when this is used as a 3-D scanner, providing a useful guide, for researchers and practitioners, in an informed choice of the optimal device for their own RE application.