Kategorie: ‘ROS’
Automated robotic dismantling/disassembly – screw detection and removal for electric vehicle recycling
Fully automatic disassembly with precision!
Our new robotic system is revolutionizing the disassembly of electric car batteries. Using the Neura Lara 8 robot and the latest image processing technologies, we recognize screws fully automatically and position the robot precisely to remove them safely.
Thanks to the integration of YOLOv8 and Intel RealSense depth cameras, the system can locate screws in real time and position them optimally on its own. No manual intervention necessary – the system works completely autonomously!
Our goal: to make the recycling process safer, faster and more efficient. Fewer risks for workers and maximizing the recovery of raw materials at the same time. This is the future of the circular economy!
You can find more information about #dimonta here.
contact:
Markus Schmitz
Daniel Gossen
Semi-Automated Tile-Laying Aid
Der an dieser Stelle eingebundene Inhalt führt Sie auf Seiten, die von der von Google betriebenen Seite YouTube - YouTube, LLC, 901 Cherry Ave., San Bruno, CA 94066, USA - zur Verfügung gestellt werden. Mit dem Aufruf des Inhalts kann YouTube Ihre IP-Adresse und die Sprache des Systems, sowie verschiedene browserspezifische Angaben ermitteln. Wenn Sie in Ihrem YouTube-Account eingeloggt sind, ermöglichen Sie YouTube, Ihr Surfverhalten direkt Ihrem persönlichen Profil zuzuordnen. Dies können Sie verhindern, indem Sie sich aus Ihrem YouTube-Account ausloggen. YouTube verwendet Cookies und Tracking-Tools. Die Datenverarbeitungsvorgänge sowie die Zwecke der Verarbeitung können direkt bei YouTube erfragt und eingesehen werden.
As part of the ErgoFli project, an innovative system is being developed in collaboration with project partners to help tile layers make their work more ergonomic and efficient.
In the video you can see how the system works in the Gazebo simulation environment. The robot automatically removes tiles from a magazine and matches them perfectly to the tiles already laid. With automatic processes, several tiles can be laid in succession to optimize the work process.
Our aim is to create a tool that not only reduces the workload for tilers, but also improves their working environment. We are excited about the progress and look forward to sharing more insights with you soon!
Find out more about the project here.
Contact person:
Mark Witte
Jan Wiartalla
Developments with Trapo GmbH
The IGMR – RWTH Aachen supports Trapo GmbH in the further development and integration of trajectory planning and trajectory execution algorithms for the new Trapo Loading System robot. Compared to the old TLS, the navigation of the new TLS is realized with the MoveBase Framework of the Robot Operating System (ROS). For this purpose, together with Trapo GmbH, we have created a special configuration for car-like structure for the local MoveBase Planner. The new TLS was last presented at the LogiMat trade fair.
Contact:
Daniel Gossen
Interactive learning with ROS: The new course in RWTHJuypter
We are pleased to present our latest ROS course, which is integrated as JupyterHub under RWTHJupyter. This course offers an interactive way to learn and deepen the basics of ROS (Robot Operating System).
ROS is an open source framework that allows developers to simplify the control of robots and their sensors. It provides a set of tools and libraries that make it possible to simplify the communication between the different components of a robot and speed up the development of applications.
Our course covers the main topics required to work with ROS. These include using ROS nodes, topics, services and actions, using ROS libraries such as tf (transform library) and moveit (motion planning library), and developing your own ROS packages. Working with different robot platforms is also covered.
Special emphasis is placed on the use of Jupyter notebooks to make the course interactive and vivid. This allows students to complete the examples and exercises directly in the course environment and see the results immediately.
We would like to thank Quantstack for their support in developing the course and their developments on JupyROS. By using Jupyter notebooks and JupyROS, we have been able to create an intuitive and user-friendly environment that makes learning ROS quick and easy.
We hope that our ROS course will be useful for both beginners and advanced developers and look forward to welcoming many participants.
Contact: Markus Schmitz
Review Summerschools ROBO and ROCNI 2022
The Summerschool serves to educate and deepen the knowledge of interested students. The Summerschool ROBO and ROCNI are offered by the International Academy (IntAc) and conducted by the Institute of Mechanism Theory, Machine Dynamics and Robotics (IGMR) at RWTH Aachen University. The theoretical basics are esxtended by the practical application of simulations and on the real robot in the laboratory of the IGMR.
Contact:
Presentation of the “ErgoFli” project at the KT Colloquium 2022
The aim of the ZIM-funded project “ErgoFli” is to develop an intelligent, semi-automatic tiling aid. Its intuitive user interface, automated handling of tiles and tile adhesive, and intelligent control system are expected to reduce non-ergonomic postures by up to 66%. The ErgoFli can be moved with a motorised drive and uses linear laying kinematics to lay tiles without direct wall contact using the buttering-floating method. Tiles in two formats, 30x60 cm or 60x60 cm, can be laid in a cross bond. The robot is equipped with a tile and tile adhesive reservoir so that it can be used for at least 30 minutes without refilling and can lay 12 m2 per hour while maintaining the required standard flatness tolerances.
Using a digital twin, an early prototype has already been tested in various scenarios. While the prototype is being designed in standard CAD software, the actual model is being integrated into the Gazebo simulation environment in parallel. This enables continuous integration and testing of control system software packages in application-related scenarios. For example, real test data can be transferred to the digital twin via a closed-loop simulation. The three laser profile sensors of the tile gripper are attached to a prototype gripper and mounted on a UR10 robot arm. The measured values from the sensors can then be transferred to the digital twin in the simulation, despite the different kinematics. In this way, the position control of the gripper can be integrated into the overall system at an early stage by measuring the joint widths.
Contacts:
Jan Wiartalla
Marius Gürtler
First movement of PARAGRIP with new control unit
PARAGRIP’s control architecture has been completely redesigned in order to be fully equipped for its future tasks in multidirectional additive manufacturing (MDAM) with arc welding (WAAM). The movements of all four arms can now be planned, simulated and executed on the real robot using MoveIt in ROS2. The joint positions of the physical robot are always fed back to ROS2, allowing the integration of online planning algorithms in the future. The video shows the planning and execution of a simple test motion of the PARAGRIP.
More information about the project can be found here.
Contact: Jan Wiartalla
A multi-layered task sequencing approach
Cobots are highly sought by manufacturing companies in contrast to fully automated production lines, as they provide the additional benefit of flexible operations. A major hurdle with current collaborative setups is tedious setup times for efficient and robust co-working as well as poor support for random interruptions.
This project focuses on enabling autonomous collaborative operations for serial manipulators where interruptions from human agents occur at random while ensuring minimal setup times. To this end, two primary aspects that impact task execution are addressed, namely execution time and co-working as enumerated below:
- A method is developed to minimise the total distance travelled, by following the most optimal sequence for the given task while retaining online operational capabilities
- A real-time replanning and co-working algorithm for randomly interruptive environments is developed and implemented to ensure continued operation even when regions of the workspace are occluded while guaranteeing safety of the human agents in the workspace. The co-working controller operates fully autonomously.
An example of the working of the deployed on a prototype platform consisting of a collaborative UR10e arm, a stereo camera for static environment mapping and a laser scanner for mapping of dynamic obstacles is shown in the video.
Contact: Daniel Gossen
music: madiRFAN – Both of Us (https://pixabay.com/music/beats-madirfan-both-of-us-14037/)
Watch the video on our Youtube channel: here.
Participation at European Robotics Forum (ERF) Hackathon 2022
As part of the European Robotics Forum (ERF) Hackathon 2022, six IGMR students demonstrated their talent in prototyping and working with robots. The hackathon challenge was set by Lely, among others. Their ‘Juno‘ mobile robots are autonomous cylindrical platforms whose main task is to move the fencing around cows on farms to make the feed pushed into the space accessible again. The hackathon task was similar: two Juno robots had to move along the walls of two interconnected rooms at a given distance. Additional restrictions and challenges were added for extra points. The team was successful at the hackathon in Rotterdam. After a neck-and-neck race, the first place went to TU Delft. At the award ceremony, we were praised as the most cooperative team for our “constant support of other teams in design and 3D printing“. This ‘exemplary behaviour‘ is much appreciated and we look forward to taking part again next year.
Special praise goes to our students Sebastian Polzin, Frederik van Kerkom, Jonas Braun, Oleksander Kutovyi, Ali Berger and Yannik Freischlad for their efforts. We congratulate TU Delft on their well-deserved victory and are happy to have won many new friends and valuable contacts. We thank the institute management for the opportunity to participate and look forward to next year.
Contact:
Integrated AI task planning and screw recognition in a production scenario
Der an dieser Stelle eingebundene Inhalt führt Sie auf Seiten, die von der von Google betriebenen Seite YouTube - YouTube, LLC, 901 Cherry Ave., San Bruno, CA 94066, USA - zur Verfügung gestellt werden. Mit dem Aufruf des Inhalts kann YouTube Ihre IP-Adresse und die Sprache des Systems, sowie verschiedene browserspezifische Angaben ermitteln. Wenn Sie in Ihrem YouTube-Account eingeloggt sind, ermöglichen Sie YouTube, Ihr Surfverhalten direkt Ihrem persönlichen Profil zuzuordnen. Dies können Sie verhindern, indem Sie sich aus Ihrem YouTube-Account ausloggen. YouTube verwendet Cookies und Tracking-Tools. Die Datenverarbeitungsvorgänge sowie die Zwecke der Verarbeitung können direkt bei YouTube erfragt und eingesehen werden.
The video is also available on our YouTube channel.
Contact:
Daniel Gossen
