Riemann project, ROS-based Education of Advanced Motion Planning and Control

Motivation:

The European warehouse robotics market is expected to grow at over 10 % in 2020 - 2025. One component of warehouse automation is fleet-deployment and mobile manipulation for autonomous tasks. Many companies in this field utilising robotic systems face a lack of knowledge to configure or modify the techniques by themselves. Especially for open-source-based robotic systems or libraries, a professional education environment is often not established yet. The diversity of employees that should be taught is usually not regarded in technical documentation and material.

Objective:

This project aims to reduce technological barriers towards using a fleet of robots in warehouses and manufacturing environments. Upskilling their employees will reduce the prototyping time for companies manufacturing robots or related software services. Our education materials will help industries use mobile robot solutions to perform complex debugging/maintenance without overly relying on their third-party supplier.

Approach:

This project creates learning material to teach skilled professionals in advanced autonomous navigation concepts, specifically how to deploy open-source software libraries on different mobile robot platforms. These advanced yet-didactic educational materials for the EIT educational platform will enable the next step and help end-users and professionals understand the inner workings of robotic systems and different navigation algorithms implemented in the Robot Operating System (ROS) and other open-source frameworks. The courses target professionals of all groups and skill levels. By building on fundamental courses on ROS and robotic systems, the audience learns at a speed fitting their needs with regular assessments to measure their current level. Learners can also skip the theoretical background to only focus on necessary implementation skills if needed. This design will save time spent on unnecessary training due to ill-fitting coursework. Some learning material is readily available on the Eupopean online reskilling platform Skills.move and the content developed within the RIEMANN 2023 project will be available upon completion but can be made accessible upon request.

Facts:

RIEMANN 2022

Project name	ROS-based Education of Advanced Motion Planning and Control
Project acronym	RIEMANN
Activity ID	22235
Duration	2022-01-01 to 2022-12-31
Program	Horizon Europe, EIT Manufacturing Education, call 2022

RIEMANN 2023

Project name	ROS-based Education of Advanced Motion Planning and Control
Project acronym	RIEMANN
Activity ID	23517
Duration	2023-01-01 to 2023-12-31
Program	Horizon Europe, EIT Manufacturing Education, call 2023

Topics & Learning Paths

RIEMANN 2022

The learning material developed during RIEMANN 2022 is already available to the public on the Skills.move platform.

Topics:

Robot Operating System: from ROS1 to ROS2
An overview of robotic planners and controllers
Introduction to ROS2 Navigation Stack
Robot Motion Model for Wheeled Robots
Environment setup for ROS2
Simultaneous Localisation And Mapping
Practical application of ROS2 Navigation Stack on a mobile robot
Reactive Collision Avoidance: Dynamic Window Approach
Rapidly-Exploring Random Tree
Implementation of a ROS2/Navigation2 plugin: RRT algorithm as global planner
Types of Distributed Control for mobile robotics
Frameworks for Distributed Control of autonomous mobile Robots in ROS
Model Predictive Control - An introduction
Model Predictive Control - How to implement?
Model Predictive Control - The Main Ingredients
Artificial Potential Fields
Artificial Potential Fields - Setup
Timed Elastic Band Planner - Part 1
Timed Elastic Band Planner - Part 2

Learning Paths:

The partners agreed on seven learning paths for the different developed topics.

Introduction to ROS2
Navigation Stack in ROS2
Dynamic Window Approach
Distributed Control
Model Predictive Control
Artificial Potential Fields
Timed Elastic Band

RIEMANN 2023

The learning material of the RIEMANN 2023 is currently being written and is subject to changes. Some of the learning material is however already available and can be accessed, either in the form of in-presence workshop or online, by contacting one of the project partners.

Topics:

Kinematic model description
Inverse Kinematics
Collision Geometry and collision-free inverse kinematics
Motion planning; Basic Concepts
Reactive Motion Planning
Off-the-shelf Motion Planners in ROS
Trajectory Optimization: Basic Concepts
Trajectory Optimization Libraries integrated within ROS eco-system
What is hybrid planning for robotic manipulators?
Global planning for robotic manipulators
Local Planning for robotic manipulators
Capabilities, limits of simulators for robotics
Differences to the real robot
e.Do Cube in Gazebo
Checking robot collision with pyBullet
Introduction to MoveIt!
Universal Robot Description Format (URDF)
MoveIt! Configuration file, SRDF
MoveIt Setup Assistant in ROS 2
Configuring ros2_control for your robot in ROS 2
Pipeline configuration for MoveIt!
ROS 2 launch system
How to setup e.DO into ROS workbench for FreeCAD
Configure, calibrate and basic movements of e.DO
Integration into move_group interface (ROS 2 integration)
DDS/RTPS – What is it?
ROS Middleware implementations – overview
Working with different RMW
An interface between e.DO Cube and Turtlebot3 using a ROS Middleware Frame
Managing your message throughput with Quality-of-Service settings in ROS 2
Using shared memory to reduce latency and computing power in ROS 2
Hierarchy and architecture of multi-agent systems in ROS 2
Example of fleet management with ChoiRbot
Spawning multiple robots in gazebo simulator; initializing individual global and local planner.
Integrating off-the-shelf planners on multiple robots in ROS 1 Gazebo: local and global path
Integrating a custom planner: APF on multiple robots
Installation of the ROS workbench for FreeCAD
Definition of a manipulator in the ROS workbench
Definition of a mobile robot in the ROS workbench
Export to URDF from the ROS workbench
Describing the robot system for a mobile manipulator with xacro
MoveIt!2 for manipulators
Using MoveIt!2 for mobile manipulation
Behaviour-Trees

Learning Paths:

Manipulator Motion Planning Part I
Manipulator Motion Planning Part II: position-controlled robots
Manipulator Motion Planning Part III: Advanced Motion Planning for Manipulators
Hybrid Planning for manipulators
Basics of manipulator simulators
Configuring your robot manipulator for ROS 2
Configure e.Do Cube
ROS2 Middleware implementations
Fleet management in ROS 2
ROS-based multi-robot planning
Set up your robot for ROS 2 with the ROS Workbench for FreeCAD
Mobile Manipulation

Project Partners

The "BIBA - Bremer Institut für Produktion und Logistik" is a research institute focused on engineering science. It was founded in 1981 as one of the first affiliated institutes of the University of Bremen and counts as one of the largest research facilities in the federal state of Bremen. It consists of two divisions: "Intelligent Production and Logistics Systems", headed by Prof. Dr.-Ing. Michael Freitag and "ICT applications for Production", headed by Prof. Dr.-Ing. Klaus-Dieter Thoben. Based on distinct fundamental research, BIBA carries out applied and contract research, both on national and international level, in the areas of production and logistics for relevant industries like logistics, automotive, aviation and wind energy. Furthermore, BIBA is involved in a University of Bremen-based Collaborative Research Center, funded by the German Research Foundation, as well as in the Bremen Research Cluster for Dynamics in Logistics (LogDynamics) and the International Graduate School for Dynamics in Logistics. With the LogDynamics Lab, the institute operates a demonstration and application center for innovative mobile technologies in logistics.

The Czech Institute of Informatics, Robotics and Cybernetics is a modern research institute of the Czech Technical University (CIIRC CTU), which brings together excellent research teams, young talent and unique know-how in order to push the technological boundaries and build on the best of Czech technical education traditions. CIIRC CTU's research focuses on four fundamental pillars: industry, energy, smart cities and a healthy society, in both basic and applied research. Technology transfer from academia to practice is an important commitment for us. CIIRC CTU is a dynamically developing and growing institution that in 2022 employs more than 300 employees, mainly researchers.

University of Tartu Institute of Technology (TUIT) is a research and development institution established in 2001 and is part of the Faculty of Science and Technology. Our goal is to lay the foundation for novel technological solutions and therefore be the leader of innovation. Institute of Technology is a multidisciplinary research institution that gives broad-based education. It gives a strong advantage in Estonian and also in the world job market. Institute unites computer engineering, robotics, electronics, material science, biosciences and environmental technologies, therefore our students have a broad opportunity to work with scientists who are creating novel solutions in all fields. Practical courses give students the opportunity to gain necessary skills in order to work in industry during and after the studies. The scientific work of our Institute focuses on four main areas - Biomedical Technologies, Environmental Technologies, Systems and Synthetic Biology, Intelligent Materials and Systems, Intelligent Computer Vision.

Comau, a member of the FCA Group, is a worldwide leader in delivering advanced industrial automation products and systems. Its portfolio includes technology and systems for electric, hybrid and traditional vehicle manufacturing, industrial robots, collaborative and wearable robotics, autonomous logistics, dedicated machining centers and interconnected digital services and products. Headquartered in Turin, Italy, Comau has an international network of 32 locations, 14 manufacturing plants and 5 innovation centers that span 14 countries and employ more than 9,000 people.

Riemann project, ROS-based Education of Advanced Motion Planning and Control

Motivation:

Objective:

Approach:

Facts:

RIEMANN 2022

RIEMANN 2023

Topics & Learning Paths

RIEMANN 2022

Topics:

Learning Paths:

RIEMANN 2023

Topics:

Learning Paths:

Project Partners

Contact