A Flexible Robotic Framework for Autonomous Manufacturing Processes
The goal of the project is to provide a basic framework for intelligent Pick-and-Place tasks.
It is based on ROS and makes use of state-of-the-art algorithms and libraries.
In 2014, the European Robotics Challenges (EuRoC) were launched with the aim of advancing robotic technologies for the next generation of intelligent industrial processes.
The Institute of Applied Mechanics of the Technical University Munich participated in the challenge 2 of the EuRoC with the team AM-Robotics. In it, competitor teams had to develop autonomous robot systems for logistics and robotic co-workers.
For the stage 1 (simulatio), various pick-and-place tasks had to be performed without any user interaction while dealing with uncertain object localization, calibration errors, obstructed environments and multi-level decision processes.
The team AM-Robotics ranked first at stage 1 of the challenge amongst 39 teams, using this framework. It includes modules for object recognition and environment modeling, grasping, motion planning as well as an intelligent state machine.
Videos of the framework in action can be seen on youtube:
EuRoC first round (simulation - Challenge 2) AM-Robotics
EuRoC AMRobotics mixed simulations
And in the following publications a detailed description can be found:
The project is published under the terms of the CRAPL License.