Skip to main content

Co-production CeLL performing Human-Robot Collaborative AssEmbly

Periodic Reporting for period 2 - CoLLaboratE (Co-production CeLL performing Human-Robot Collaborative AssEmbly)

Reporting period: 2020-04-01 to 2021-09-30

The CoLLaboratE is a 42-month project that aims to enable genuine human–robot collaboration for performing assembly tasks in a production cell that is designed to provide a safe collaborative environment. The project builds upon state-of-the-art methods for teaching the robot assembly tasks using different means of human demonstration, equips the robots and AGV mobile platforms with basic collaboration skills and provides efficient safety strategies for a fenceless approach within the production cell. As a result, closer collaboration will be achievable and efficient production plans making optimal use of the available resources can be designed and executed.
For CoLLaboratE to successfully realize its vision, a few scientific and technological objectives have been set throughout the project duration. These are listed in the following points:
1) To equip the robotic agents with basic collaboration skills easily adaptable to specific tasks
2) To develop a framework that enables non-experts teaching human-robot collaborative tasks from demonstration
3) The development of technologies that will enable autonomous assembly policy learning and policy improvement
4) To develop advanced safety strategies allowing effective human robot cooperation with no barriers and ergonomic performance monitoring
5) To develop techniques for controlling the production line while making optimal use of the resources by generating efficient production plans, employing reconfigurable hardware design, and utilising AGV’s with increased autonomy
6) To investigate the impact of Human-Robot Collaboration to the workers’ job satisfaction, as well as test easily applicable interventions in order to increase trust, satisfaction and performance
7) To validate CoLLaboratE system’s ability to facilitate genuine collaboration between robots and humans
In the second reporting period of CoLLaboratE, the consortium continued their efforts in accordance to the work-plan to finalize all the modules needed for collaborative assembly and to start the integration efforts for the deployment of the four use-cases.

The achievements in this second reporting period include:
- Amendments and contingency measures to handle the delays of the COVID pandemic, while turning all meeting to virtual ones (WP1)
- Finalization of the user-requirements and system architecture, while defining the use-cases to facilitate integration (WP2)
- Finalization of the basic collaboration skills for efficient, adaptive and safe interaction (WP3),
- Finalization of the different modalities to teach assembly tasks by demonstration to a robot (WP4)
- Finalization of the high-level solutions for monitoring ergonomics, interfacing with the robot manipulators and the AGVs and overall task planning (WP5)
- Remote integration of the CoLLaboratE modules in most of the use-cases (WP6), to the extent that it was possible, due to COVID-19
- A web focused communication & dissemination strategy due to the travel restrictions
In the second reporting period, the CoLLaboratE partners have produced about 30 open-access scientific publications in international conferences and journals that are all available through the projects website.
The progress made by the project beyond the state of the art is highlighted in the following achievements:
- A method for variable admittance controller to allow easy and precise manipulation of large objects with high inertia.
- A method for high tracking accuracy in following a desired trajectory under model and task uncertainties utilizing a highly compliant robot, ensuring safety under unintentional contact.
- A novel Dynamic Movement Primitive formulation that supports backward reproduction of a learned trajectory while allowing two stage learning.
- Several behavior primitives such as search patterns, manipulability behaviors, impedance behaviors and reactive behaviors, which can be used as bricks to create more complex movements.
- A framework where the initial demonstration evolves into a natural physical human-robot collaboration. The core of the proposed approach is a phase feedback loop from the robot to DMPs and compliance along the motion trajectory.
- A key-frame extraction framework that performs semantic analysis on videos of human demonstrations, to address collaborative assembly tasks.
- An incremental kinesthetic guidance framework, which allows easy and natural editing of robotic skills, based on reversible DMPs.
- An intuitive and easy-to-use mobile interface which relies on augmented reality to teach robot paths with variations.
- An extension to active learning strategy to improve trajectory policies which can measure the uncertainty over the whole path instead of a single state.
- A robot control scheme that enforces the dynamic active constraints produced by a vision system, protecting its environment, including humans, from full robot body collisions.
- A framework for human intention recognition in collaborative tasks based on recurrent neural networks.
- A production planner that enables the allocation of tasks to the available resources of a production line.
- Ergonomic monitoring module that models human dynamics and detect patterns in the motion trajectories that imply exposure to an ergonomic risk factor.
- A gesture recognition module that permits to the collaborative robot to adapt its actions spatiotemporally reaching thus a more natural collaboration.
- An optimization procedure for optimal layout and configuration of the reconfigurable fixtures in the work-cell, taking into account the ergonomics regarding the human co-worker.
- A study to identify factors that influence job satisfaction and quality of the workers of the companies participating in the use cases, and a questionnaire measuring the identified factors.

The expected potential impact from the project, as it has been identified by the potential end-users, is highlighted as follows:
- Introduction of human-robot collaboration on continuous production lines and tasks that were previously deemed not appropriate for robot automation
- Ergonomic improvement, particularly in intensive tasks performed manually such as riveting.
- Easy integration in production, drastically reducing the cost of programming.
- Versatility and adaptability, as the robot can be easily taught various collaborative activities.
Use-case 2: Windshield assembly and inspection
Use-case 1: Collaborative car starter assembly setup
Use-case 3: TV assembly
Use-case 2: Setup for windshield assembly and inspection
Use-case 1: Car starter assembly
Use-case 4: Aircraft structure riveting
Use-case 4: Integrated mobile robot arm for aircraft structure riveting
Use-case 3: Collaborative TV assembly setup