Community Research and Development Information Service - CORDIS

H2020

SYMBIO-TIC Report Summary

Project ID: 637107
Funded under: H2020-EU.2.1.5.1.

Periodic Reporting for period 1 - SYMBIO-TIC (Symbiotic Human-Robot Collaborative Assembly: Technologies, Innovations and Competitiveness)

Reporting period: 2015-04-01 to 2016-03-31

Summary of the context and overall objectives of the project

The European robotics industry is moving towards a new generation of robots, based on workplace safety and the ability to work alongside with humans. This new generation is paramount to making the factories of the future more cost-effective and restoring the competitiveness of the European manufacturing industry.

However, the European manufacturing industry is facing the following challenges:
• Lack of adaptability – dynamic changes or deviations on shop floors require real-time monitoring and adaptive execution control with smart sensor networks for context-aware information sharing and task planning as quickly as possible;
• Lack of flexibility – manufacturing complexity and dynamism demands responsive human-robot interactions to address increasing diversity and speciality of production equipment and processes in difficult working environment as flexibly as possible; and
• Lack of vertical integration – today’s global competition pushes for much quicker time-to-market by vertical integration between task-planning computers and task-execution robot controllers with zero tedious programming as seamlessly as possible to increase manufacturing throughput.

The SYMBIO-TIC project addresses these important issues towards a safe, dynamic, intuitive and cost-effective working environment: immersive and symbiotic collaboration between human workers and robots can improve this situation and bring significant benefits to robot-reluctant industries where current tasks and processes are perceived to be too complex to be automated. The benefits include lower costs, increased safety, better working conditions and higher profitability through improved adaptability, flexibility, performance and seamless integration.

This project is addressing the topic H2020-FoF-6-2014, aiming at a novel hybrid assembly/packaging ecosystem in dynamic factory environment based on human-robot collaboration. The system is context-aware in task planning and execution, safe to human workers in a shared fenceless working space, flexible and adaptive to dynamic changes, and cost effective. The focus of this project is centred on innovative technologies for human-robot collaboration that is multimodal, user-friendly (programming-free to end users) and replicable to real-world industrial applications. The project (1) bridges the gap between automated but inflexible robotic tasks and labour-intensive manual operations; (2) combines the power and repeatability of robots with the accuracy and flexibility of humans; (3) improves the overall productivity through reliable and safe human-robot collaboration; and (4) enhances the system adaptability against unforeseen changes in production.

The ultimate goal of this project is to contribute with innovative technologies to factories of the future where European manufacturers can compete effectively in the global market. In order to implement such a human-robot tight collaborative system, four challenges must be addressed: (1) how to safeguard human workers at all time when interacting with robots in a shared fenceless environment; (2) how to generate who-do-what work plans on the fly suitable for the human-robot mixed tasks; (3) how to adapt dynamic changes and control robots quickly and correctly with zero programming for robot users; and (4) how to interface with robots via multimodal interfaces efficiently as well as to instruct human workers on what-to-do and/or how-to-do effectively, especially for human-robot complex assembly/packaging tasks.

The above 3 key areas and 4 challenges are addressed in this project through innovative technologies based on smart sensors (visual, auditory, haptic, etc.), and self-adaptive decision algorithms. A human-robot collaborative assembly/packaging system is chosen as a test-bed for testing and validation in semi-automated working environment with human involvement. The aforementioned 4 challenges are mapped to 4 specific objectives below to be achieved collectively by the project partners in 4 years. The fifth specific objective is to validate the SYMBIO-TIC outcomes and their scalability through 3 physical demonstrators of varying scenarios in 3 industrial sectors: food-processing, aeronautic and automotive.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

In the reporting period, the SYMBIO-TIC project has established the basis of the following objectives:

Objective 1: To develop an active collision avoidance subsystem to safeguard human workers
This objective is to guarantee the safety of the workers in a shared human-robot fenceless immersive environment. It addresses the call scope in the areas of “innovative strategies for online safety monitoring” and “safety during a mechanical failure of the robotic system”. It contributes to symbiotic collaboration through “intuitive and multimodal programming to allow robot systems to be rapidly and easily programmed without prior knowledge on robot systems”. Although research on collision detection has been active for many years, it is still impractical in industrial settings when human workers are purposely exposed to robots during tight collaboration. When safety is concerned, timing becomes critical. A target robot reaction latency of less than 50 msec (much quicker than human movements) is therefore set and anticipated in this project. This objective ensures a shared fenceless working space for human workers and robots, and is supported by WP1.

Objective 2: To generate adaptive task plans appropriate to both robots and human workers
This objective aims at the future factories with immersive collaboration between human workers and robots by developing “new methodologies for the initial planning and online dynamic re-planning of the shared tasks”. For a human-robot co-existent and collaborative system, each party should better do what he/she/it is good at when accuracy, flexibility, repeatability, strength and ergonomics are taken into account. In other words, an adaptive task plan should consider improved productivity of shared tasks in shared workspace by selecting the right worker or robot for the right task. Reducing energy consumption is also addressed in this objective when planning robot trajectories that can perform the planned tasks using the least energy for better cost-effectiveness, higher profitability and long-term sustainability. This objective targets the task sharing between human workers and robots where planned tasks must be shared or even switched dynamically due to the current availability of human workers and robots, or when an alternative becomes the only option at the moment for a particular task. It thus addresses the organisational issue of difficult working environment. Within the project, this objective is supported by WP2.

Objective 3: To adapt to dynamic changes with intuitive and multimodal programming
The call requires human-robot interactions through “intuitive and multimodal programming to allow robot systems to be rapidly and easily programmed without prior knowledge on robot systems”. This objective is aligned with this requirement using multimodal programming methods in complex shared robotic tasks, resulting in zero tedious programming for robot users. Instead of specifying how-to-do at the early planning stage, this objective empowers robot controllers to decide how-to-do at the task execution stage based on the what-to-do list of a high-level task plan (the outcome of Objective 2) and the safety policy applied. It also generates native robot control commands based on multimodal human instructions, resulting in a novel robot programming method that is multimodal, intuitive, fast and adaptive. It targets a totally programming-free environment for workers on shop floors. This objective is supported by WP3.

Objective 4: To provide human workers with in-situ assistance on what-to-do and how-to-do
This objective is to support walking workers in an immersive and collaborative human-robot assembly environment, who need in-situ assistances (what-to-do and how-to-do due to location and task differences). Therefore, location-aware task-specific assistances to any worker are the main focus of this objective. Such assistance includes (1) multimodal interfaces with robots (haptic interaction via force or torque, robot motion guidance by gestures or auditory commands, etc.), and (2) multimodal devices for the human workers (3D goggles, headsets, arm-band devices). Making the most out from employees’ knowledge and skills based on a competence matrix of the human workers is another organisational issue to be addressed in this objective. This objective is supported by WP4.

Objective 5: To demonstrate and validate the project concept and solutions
This objective is to test and quantify the aforementioned 4 objectives in terms of safety, feasibility, intuitiveness, adaptability and scalability of collaboration between humans and robots. A test-bed integrating the outcomes of other objectives will be developed to showcase the envisioned technologies, innovations and competitiveness in this project. It addresses the topic of “safety during a mechanical failure of the robotic system during tight collaboration of humans and robots” by validating the safety policies, active collision avoidance subsystem and embedded robot control algorithms. 3 demonstrators placed in 3 European Member States are planned, based on the test-bed, to demonstrate project solutions in food-processing, aeronautic and automotive industries, respectively. The three demonstrators have been pre-selected based on the following criteria: (1) level of complexity – from easy food-processing to complex automotive assembly, and (2) scope of coverage – from robot-reluctant food-processing industry to robot-reluctant engine assembly processes. Although automotive industry is considered robot-friendly with highly robotised spot-welding and painting operations, engines are still assembled by hand over long stretches due to their complex structure with many levels and individual parts that must be assembled in a correspondingly precise way. The engine assembly is thought difficult to automate and thus robot-reluctant processes in a robot-friendly industry. Linked to WP5, this objective aims at validation in near real industrial settings (onsite but non-production environment) at participating partners’ facilities focusing on different scenarios at different levels of complexity:
− A food-packaging demonstrator by robomotion GmbH in Germany (a system integrator);
− An aeronautic component assembly demonstrator by Aciturri in Spain (an end-user); and
− An automotive engine assembly demonstrator by Volvo Car Corporation in Sweden (an end-user).

The results of the validations are used for continuous improvement of the solutions and for identifying new standards that may be required for future scale-up towards industrial applications in real production environment, if not yet governed by existing national/international standards. Stable demonstrators will be used in dissemination and exploitation activities, including 3 industry-oriented workshops.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The project industrial impact is supported by large industrial partners (VCC, ABB, ACI, IDK) and SMEs (PRO, PRD, SAN, AMS, ROB) that will allocate substantial resources to market dissemination (WP6) and exploitation (WP7) activities (please refer to Section 2.2).
SMEs form the EU economy’s industrial backbone; however their productivity lags 30% behind large enterprises. In industrial manufacturing, a part of the lead of the large enterprises is the productivity increase generated by robots.
The project will support widespread adoption of robots by SMEs resulting in productivity improvements and therefore will provide a competitive advantage to the European economy as a whole. This competitive advantage will support the re-shoring of industrial activities to Europe and contribute to Europe’s job creation and economic recovery: the technology developed in the project will contribute a rebalancing of world manufacturing economics, enabling traditionally higher labour rate countries to compete in world markets. Greater competiveness results in increased sales of manufactured products leading to increased number of manufacturing jobs and creates higher paying support jobs.
This project will also contribute to the development of a regulatory framework around human-robot collaborative systems (please refer to Section 2.1.5) and provide input to European Technologies Platforms and PPPs such as EURobotics PPP (please refer to Section 2.1.6).

The project scientific impact is supported by the academic and research partners (KTH, MTA, LMS, HIS, VTT, IPA) and rests on a rich set of scientific dissemination activities (please refer to Section 2.2). This project will influence the robotics research community by focusing attention on human-robot interaction and fenceless environments (and away from traditional safeguarding and clearance aspects). Specifically, it is expected that the project key innovations (programming-free methodology, safety policy-guided sensor-driven task planning/re-planning, active collision avoidance supported by sensor-driven 3D models) will have a durable impact on robotics hardware and software research. The project will also influence the legal framework of human-robot workplace collaboration. As current standards are insufficient (ISO and safe-regulation norms do not have legal character), compliance with safety measures does not provide protection from legal risk. Also, the more autonomous a robot is, the more its actions are unpredictable, which raises concerns about the foreseeability of the robot’s behaviour in certain situations and dangers arising from it. Human-robot collaboration will therefore require new legal options for employees (refusal to work or right for hazard pay), new methods for the protection of workers and in general, new regulations.

The project societal impact results from addressing specific challenges such as health & safety (H2020 societal challenge No. 1) and resource conservation (H2020 societal challenge No. 5).
Many production jobs in manufacturing involve repetitive, physically demanding work, which could be effectively fulfilled by robots. Workers are highly susceptible to repetitive-strain injuries to their hands, wrists, and elbows . Production workers often stand for long periods and may be required to lift heavy objects or use cutting, slicing, grinding, and other dangerous tools and machines. To deal with difficult working conditions and comply with safety regulations, companies have initiated ergonomic programs to cut down on work-related accidents and injuries. This project will demonstrate that the development of lower cost, safer and more flexible automation technologies such as collaborative robots can lower worker injuries while increasing productivity and cost-effectiveness.
Robots have the capability to perform tasks more accurately than their human counterparts, therefore reducing waste in the manufacturing process (more efficient use of feedstock, lower rejection rate of manufactured parts). For example, high-pressure painting using industrial robots has enabled office and kitchen furniture manufacturer Spartan (an ABB customer in Slovakia) to reduce the amount of paint they use by 15 percent. Their goal was to meet the increased quality requirements of their clients, to become more environmentally friendly and to boost production capacity. Similarly, this project will demonstrate the sustainability impact of automation in industries that have traditionally had to relying on manual processes such as aeronautics.

Related information

Record Number: 190288 / Last updated on: 2016-11-14
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top