Dr. Maximo A. Roa
(DLR)
Title: Autonomous assembly for lot size one production
Time: 9:15 – 9:45
Abstract:
This talk discusses the challenges and potential solutions for automatic assembly of small lot sizes. The implementation consists of a dual arm robotic system that provides a fully automated pipeline for automatic assembly of aluminum profile constructions, starting with either a physical example or a CAD-input of the desired structure. The system includes an assembly sequence planner integrated with a grasp planning tool, a knowledge-based reasoning method, a skill-based code generation, and an error tolerant execution engine. Motion planning is also sped up by using learning methods that simplify the planning of trajectories in task and configuration space. Finally, an outlook is provided for the extension of such a system for in-space assembly of large structures, such as telescopes or modular satellites.
Dr. Daolin Ma
(MIT)
Title: High-Resolution Tactile Sensing to Enable Autonomous Assembling Solutions
Time: 9:45 – 10:15
Abstract:
A great leap has been seen in computer vision during the last decade, which has made robotic automation solutions possible in the logistic industry. However, for assembly tasks, vision perception alone cannot solve the problem. Human workers rely both on eyes and hands for their amazing assembling skills. Because assembling tasks are contact-rich, a missing technology for automating them is a “better hand” that can provide contact feedback in the form of contact geometry, contact force and contact slip.
In this talk, I would introduce our recent work on development of a tactile sensor-GelSlim, for dense force estimation, autonomous packing and assembly. GelSlim is a high-resolution vision-based tactile sensor that 1) Provides the low-level tactile information; 2) Enables real-time autonomous packing and assembly. The tactile feedback makes accomplishing these tasks possible from both perception and control perspectives, which indicates that tactile feedback can be an enabler for autonomous assembling solutions.
Dr. Dong-Hyuk Lee
(Korea Institute of Industrial Technology)
Title: Object assembly using dual-arm robot and dexterous robot hands
Dong-Hyuk slides
Time: 10:15-10:45
Abstract: TBD
Prof. Christian Schlette
(University of Sothern Denmark)
Title: Towards hyper-modular robot cells for high mix/ low volume assembly
Christian slides
Time: 11:15 – 11:45
Abstract:
To support shifting to high mix/low volume production, central efforts in agile production push for robotizing all sorts of production operations – with a special focus on assembly, where any complexities resulting from part and product varieties today are still mostly countered by commissioning manual operations. Attempts to transfer these manual assembly operations to fully automated assembly cells thus need to systematically address rapidly evolving varieties with hyper-modular approaches that eventually allow for implementing similar levels of agility as manual operations. At SDU Robotics, we designed a robot cell for the World Robot Challenge / Industrial Category / Assembly Challenge 2018, with which we tried to overcome the current limitations in robotic assembly. This cell is a starting point towards the design and operation of hyper-modular robot cells for high mix/ low volume assembly. The talk will address the current limitations as well as first and future technologies for fully automated, agile robot production systems.
Prof. Tokuo Tsuji
(Kanazawa University)
Title: Simple Assembly system for quick adjustment
Time: 11:45 – 12:15
Abstract:
We develop a simple assembly system for setting it quickly. Assembly system tends to be large because it needs a lot of hands, jigs and tools. As the system grows, the cost of setting increase. At the World Robot Summit, the system is expected to be leanness and agile for being quickly rearranged. Therefore, we implement a simple system that uses only two hands. In addition, robust operations are executed by using sensors. We evaluate this system in the competition of the World Robot Summit.
Prof. Nikolaus Correll
(University of Colorado at Boulder)
Title: Robotic assembly using impedance control and in-hand 3D perception
Nikolaus slides
Time: 13:45 – 14:15
Abstract:
We describe a series of behaviors for autonomous assembly of commodity mechanical parts that include peg-in-hole and hole-on-peg insertions, screwing of screws and nuts, and handling of rubber bands under tension from the plate assembly task at the “World Robot Summit” and the Siemens Learning challenge. All tasks are executed using a two-finger gripper (Robotic Materials’ SmartHand) and a portable screwdriver. We make extensive use of both in-hand 3D perception and impedance control of the hand to detect parts and placement location in the environment as well as state transitions during the assembly process. We describe on-going challenges in perception, error detection and correction, and automatically generating assembly programs from structural information on the problem.
Dr. Ixchel G. Ramirez-Alpizar
(National Institute of Advanced Industrial Science and Technology)
Title: Towards an efficient framework for generating robotic assembly motions
Time: 14:15 – 14:45
Abstract:
Among the tasks that remain done by humans, assembly tasks are particularly difficult to execute by a robot. We propose two different methods for generating robotic assembly motions. First, I will introduce ongoing research on the development of a framework based on a virtual environment, specifically designed for recording assembly motions done by humans to be later on reproduced by a dual arm robot. Second, I will introduce our research on motion recognition for assembly tasks towards the easy reuse and efficient generation of new assembly motions for robots, aiming for a full automatization of assembly tasks.
Dr. Josie Hughes
(University of Cambridge)
Title: TBD
Time: 16:15 – 16:45
Abstract: TBD
Dr. Felix von Drigalski
(OMRON SINIC X)
Title: Robotic assembly and manipulation – a reality
Felix slides
Time: 16:45 – 17:15
Abstract:
Despite being expected to assemble complex products with small lot sizes in the very near future, robotic systems are only used in strongly controlled environments with large lot sizes. This talk presents the obstacles to fully automated assembly with current robotic systems, the requirements for vision and sensing systems, state of the art approaches to the problem, using examples from industrial practice and the World Robot Summit Assembly Challenge 2018 held in Tokyo.
Prof. Yasuyoshi Yokokohji
(Kobe University)
(Chairperson of the WRS Industrial Robotics Competition Committee)
Title: World Robot Summit (WRS) Assembly Challenge – Results of WRS 2018 and Outlook for WRS 2020
Yasuyoshi slides
Time: 17:15 – 17:45
Abstract:
The World Robot Summit (WRS) is a robotic “challenge and exposition” organized by the Japanese government to accelerate social implementation, research and development of robots working in realistic daily life, society, and industrial fields. “Assembly Challenge” is a robot competition of the Industrial Robotics Category of the WRS, which is organized by the WRS Industrial Robotics Competition Committee in order to promote the development of the next-generation production systems that can respond to new production demands in agile and lean manners. In this talk, I will give a summary of the pre-competition which was held in 2018 with 16 participating teams from around the world, followed by a perspective for the 2020 main competition.