Periodic Reporting for period 1 - CONVERGING (Social industrial collaborative environments integrating AI, Big Data and Robotics for smart manufacturing)
Okres sprawozdawczy: 2022-09-01 do 2024-02-29
Recent years have been marked by global scale occurrences including economic crisis and an unprecedented pandemic which have provided tremendous challenges on the manufacturing firms that had to rethink their production and business models. Under the new reality, reconfiguration has to consider many more dimensions than a simple rescheduling of pending orders. The production system itself needs to adopt both human and automated resources that can work together seamlessly or mutually exchange tasks, allowing the execution of any process plan in more than one, non-predetermined ways.
Thankfully, over the last decade, significant breakthroughs have been achieved on the technological enablers that can support flexible production in smart factory setups.
The CONVERGING project, as an Innovation Action, aspires to bring together a team of “Best in Class” experts that can combine all the aforementioned technological enablers to:
“Develop, deploy, validate and promote smart and reconfigurable production systems including multiple autonomous agents (collaborative robots, AGVs, humans) that are able to act in diverse production environments. The diversifying factors will be a multi-level AI based cognition (line, station, resource levels) which will exploit the collective perception (Digital Pipeline) of these resources, allowing them to interact with each other and seamlessly coexist with humans under a «social industrial environment» that ensures trustful, safe and inclusive user experience”
The proposed HRC, AI, Big Data and technologies will be tested against robustness, efficiency and acceptance in production sectors of Automotive (FORD), Aircraft Production (IAI), White Goods (ELUX) and Additive Manufacturing products processing (PRIMA)
The workplan of CONVERGING has as a target to comply with the following 5 objectives:
Objective O1: Implementing a highly reconfigurable production system by deploying collaborative robotics and smart mechatronic devices, relying on multi-level AI to achieve autonomy
Objective O2: Providing open and standard means to interconnect all production entities (Big Data pipeline) for real time capturing (Digital Twin), storing (Data at Rest) and processing (Data in Motion) to support autonomous and collaborative behaviour with minimal user intervention
Objective O3: Establishing a human centered social-industrial environment where all activities and interactions with humans are dynamically shaped to maximize user experience, trust, skills & safety
Objective O4: Providing the software and hardware interfaces to ensure safe and seamless interaction with collaborative robotic solutions, minimizing learning curves and setup times
Objective O5: Create innovation ecosystem through a network of open Pilot Lines- involving robotic application stakeholders, SMEs and RTOs to inspire further development and deployment
Thankfully, over the last decade, significant breakthroughs have been achieved on the technological enablers that can support flexible production in smart factory setups.
The CONVERGING project, as an Innovation Action, aspires to bring together a team of “Best in Class” experts that can combine all the aforementioned technological enablers to:
“Develop, deploy, validate and promote smart and reconfigurable production systems including multiple autonomous agents (collaborative robots, AGVs, humans) that are able to act in diverse production environments. The diversifying factors will be a multi-level AI based cognition (line, station, resource levels) which will exploit the collective perception (Digital Pipeline) of these resources, allowing them to interact with each other and seamlessly coexist with humans under a «social industrial environment» that ensures trustful, safe and inclusive user experience”
The proposed HRC, AI, Big Data and technologies will be tested against robustness, efficiency and acceptance in production sectors of Automotive (FORD), Aircraft Production (IAI), White Goods (ELUX) and Additive Manufacturing products processing (PRIMA)
The workplan of CONVERGING has as a target to comply with the following 5 objectives:
Objective O1: Implementing a highly reconfigurable production system by deploying collaborative robotics and smart mechatronic devices, relying on multi-level AI to achieve autonomy
Objective O2: Providing open and standard means to interconnect all production entities (Big Data pipeline) for real time capturing (Digital Twin), storing (Data at Rest) and processing (Data in Motion) to support autonomous and collaborative behaviour with minimal user intervention
Objective O3: Establishing a human centered social-industrial environment where all activities and interactions with humans are dynamically shaped to maximize user experience, trust, skills & safety
Objective O4: Providing the software and hardware interfaces to ensure safe and seamless interaction with collaborative robotic solutions, minimizing learning curves and setup times
Objective O5: Create innovation ecosystem through a network of open Pilot Lines- involving robotic application stakeholders, SMEs and RTOs to inspire further development and deployment
During the first reporting period (M01-M18), the work carried out focused around three main pillars:
i) Definition of use case scenarios and KPIs, requirements extraction and documentation, hardware and software specifications, definition of CONVERGING reference architecture
ii) Development of first prototypes of the CONVERGING enabling technologies
iii) Establishment and update of the dissemination and exploitation strategy of the project
Thus, consortium managed to achieve three Milestones (MS1, MS2, MS3), by the submission of the relevant deliverables.
The technical work carried out has led to the development of a first prototypes of the following modules:
- Dynamic Work Reorganization (DWR)
- AI Station Controller (AISC)
- Perception and Autonomy Module (PAM)
- Collaborative Robot Control (CRC)
- Humanoid Collaborative Robot (HCR)
- Remote Inspection Robot (RIR)
- Medium Payload Collaborative Manipulator (MPCM)
- Polishing Robot (PR)
- Data At Rest (DAR)
- Data In Motion (DIM)
- AI Digital Twin (AIDT)
- Operator Training Module (OTM)
- Multi-Actor Contextual Interfaces (MACI)
- User Experience and Ergonomics (UXE)
- Safety Assessment and Monitoring (SAM)
- Teaching by Demonstration Module (TDM)
- Autonomous Robot Behavior Adjustment (ARBA)
i) Definition of use case scenarios and KPIs, requirements extraction and documentation, hardware and software specifications, definition of CONVERGING reference architecture
ii) Development of first prototypes of the CONVERGING enabling technologies
iii) Establishment and update of the dissemination and exploitation strategy of the project
Thus, consortium managed to achieve three Milestones (MS1, MS2, MS3), by the submission of the relevant deliverables.
The technical work carried out has led to the development of a first prototypes of the following modules:
- Dynamic Work Reorganization (DWR)
- AI Station Controller (AISC)
- Perception and Autonomy Module (PAM)
- Collaborative Robot Control (CRC)
- Humanoid Collaborative Robot (HCR)
- Remote Inspection Robot (RIR)
- Medium Payload Collaborative Manipulator (MPCM)
- Polishing Robot (PR)
- Data At Rest (DAR)
- Data In Motion (DIM)
- AI Digital Twin (AIDT)
- Operator Training Module (OTM)
- Multi-Actor Contextual Interfaces (MACI)
- User Experience and Ergonomics (UXE)
- Safety Assessment and Monitoring (SAM)
- Teaching by Demonstration Module (TDM)
- Autonomous Robot Behavior Adjustment (ARBA)
CONVERGING has stared developing and plans to further develop a number of solutions that will enable the achievement of high impact for the EU manufacturing industry. In particular, CONVERGING supports:
- Flexible architecture: CONVERGING employs OpenFlow for data orchestration and connecting IoT devices.
- Data in Motion: It supports open interfaces and protocols for real-time data capture, including safety data.
- Data at Rest: Standardized data models are used for historical data and digital shopfloor representation.
- Human Action Recognition: Human body, hand, and speech signals are integrated for intention prediction, leveraging pre-trained networks for data efficiency.
- Digital Twins Integration: Real-time sensor data is collected and made available through DTs, enhancing shopfloor system reconfigurability and task planning.
- Digital Pipeline: A digital pipeline ensures data robustness and safety, complying with industrial requirements.
- AI Resource Autonomy: Autonomous decision-making for resource adjustment is enabled, with real-time re-planning and anomaly detection.
- HRC Station Reconfiguration: Dynamic task allocation and mission changes are based on shopfloor status, using data from safety and process sensors.
- Line and Station Reconfiguration: Autonomous robots and portable HRC solutions provide adaptability, with AI-driven task dispatching and work balancing.
- Learning and Social Autonomy: AI modules allow for skill learning and adaptation, while deep learning algorithms enable responsive HRC systems to human states.
The CONVERGING solutions have the potential to provide wider scientific, economic and societal effects, contributing to the expected impacts as follows:
- Combined Robotics and Artificial Intelligence will orchestrate a transformational effect on industry, bringing production back to EU, with benefits such as: consistency in repeated tasks, high throughput rates, customization and flexibility of production in changing market requirements, operating in unsafe environments, performing with accuracy, etc.
- CONVERGING robotic solutions will be able to solve tasks in a flexible way by making their own decisions, while the operator specifies high-level commands only.
- The CONVERGING end-users will overcome the stiffness of existing production systems by: a) Enabling existing production systems to produce at least 50% smaller lot sizes; b) Increasing 50% more product variations; c) Reducing 30% of the set-up and changeover times and costs.
- A data-driven approach based on advanced analytics and AI will allow the closed loop interaction among all the phases of the product lifecycle.
- Deep Learning will enable robots in manufacturing to deal with uncertainties, to learn behaviors through interaction with their surroundings.
- Make more attractive, upskilling and safer the jobs of the humans working in manufacturing by enabling training for integration in HRC environment.
- Flexible architecture: CONVERGING employs OpenFlow for data orchestration and connecting IoT devices.
- Data in Motion: It supports open interfaces and protocols for real-time data capture, including safety data.
- Data at Rest: Standardized data models are used for historical data and digital shopfloor representation.
- Human Action Recognition: Human body, hand, and speech signals are integrated for intention prediction, leveraging pre-trained networks for data efficiency.
- Digital Twins Integration: Real-time sensor data is collected and made available through DTs, enhancing shopfloor system reconfigurability and task planning.
- Digital Pipeline: A digital pipeline ensures data robustness and safety, complying with industrial requirements.
- AI Resource Autonomy: Autonomous decision-making for resource adjustment is enabled, with real-time re-planning and anomaly detection.
- HRC Station Reconfiguration: Dynamic task allocation and mission changes are based on shopfloor status, using data from safety and process sensors.
- Line and Station Reconfiguration: Autonomous robots and portable HRC solutions provide adaptability, with AI-driven task dispatching and work balancing.
- Learning and Social Autonomy: AI modules allow for skill learning and adaptation, while deep learning algorithms enable responsive HRC systems to human states.
The CONVERGING solutions have the potential to provide wider scientific, economic and societal effects, contributing to the expected impacts as follows:
- Combined Robotics and Artificial Intelligence will orchestrate a transformational effect on industry, bringing production back to EU, with benefits such as: consistency in repeated tasks, high throughput rates, customization and flexibility of production in changing market requirements, operating in unsafe environments, performing with accuracy, etc.
- CONVERGING robotic solutions will be able to solve tasks in a flexible way by making their own decisions, while the operator specifies high-level commands only.
- The CONVERGING end-users will overcome the stiffness of existing production systems by: a) Enabling existing production systems to produce at least 50% smaller lot sizes; b) Increasing 50% more product variations; c) Reducing 30% of the set-up and changeover times and costs.
- A data-driven approach based on advanced analytics and AI will allow the closed loop interaction among all the phases of the product lifecycle.
- Deep Learning will enable robots in manufacturing to deal with uncertainties, to learn behaviors through interaction with their surroundings.
- Make more attractive, upskilling and safer the jobs of the humans working in manufacturing by enabling training for integration in HRC environment.