Periodic Reporting for period 1 - STAR (Safe and Trusted Human Centric Artificial Intelligence in Future Manufacturing Lines) Reporting period: 2021-01-01 to 2022-06-30 Summary of the context and overall objectives of the project STAR is a joint effort of AI and digital manufacturing experts towards enabling the deployment of standard-based secure, safe reliable and trusted human centric AI systems in manufacturing environments and researches and integrates leading edge AI technologies with wide applicability in manufacturing environments, including:• Active learning (AL) systems that boost safety and accelerate the acquisition of knowledge.• Simulated reality systems that accelerate Reinforcement Learning (RL) in human robot collaboration scenarios.• Explainable AI (XAI) systems that boost the transparency of industrial systems and increase trust in them.• Human Centric digital twins enabling worker monitoring for safer and trustful production processes.• Advanced RL techniques for optimal navigation of mobile robots and for the detection of safety zones in industrial plants.• Cyber-defence mechanisms for sophisticated poisoning and evasion attacks against deep neural networks operating over industrial data.STAR’s focus on these research areas places the project at the forefront of the global research in AI in general and in digital manufacturing in particular. STAR will become a catalyst for the deployment of the most advanced AI systems in real-life manufacturing environments, through researching and providing effective ways for moving novel research results from the partners’ labs to the shop-floor. The project will leverage background projects and results of the partners in the above areas, which ensures research excellence and will enable STAR to stand out from similar research initiatives worldwide. STAR has set 8 overall objectives. These are:O1: Reference architecture and platform implementation for safe, reliable, secure and human centric AI in manufacturingO2: Transparent & XAI in manufacturing O3: Reliable AI and human-centric knowledge acquisition based on simulated reality and active learningO4: Human centred simulations and digital twins for safe AI systems in manufacturingO5: Cyber security and data reliability for AI systems in manufacturing environmentsO6: Real-life integration, validation and evaluation in production linesO7: Legal, regulatory and policy making guidelines for ethical AIO8: Market platform and digital innovation hub establishment - integration with AI4EU 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 respect to the set of eight project objectives that have been defined the work performed during this reporting period is presented below.O1: The first and second and final version of the reference architecture (STAR Reference Architecture and Blueprints-Initial version) were produced. O2: The analysis of the state of the art and potential baseline techniques for XAI has been performed, including their comparison given the diversity of AI algorithms and of the respective datasets in the manufacturing domain. O3: STAR partners followed a number of important analytical, design and implementation steps and reached among others the following progress/results: Prototype of high-fidelity data augmentation for small input datasets along with simple CNN learning model prototypes (for Simulated Reality task), simplified simulation environment for a robotic pick and place task to be used for training discrete and continuous action-space RL agents, application of different AL strategies to two use cases: (i) news classifications supporting demand forecasting in an industrial context, and (ii) defect detection for Phillips shavers manufacturing use case; O4: Progress has been made in advancing the expected AI modules for simulating and monitoring workers in production systems and supporting decision making and process optimization.O5: Work performed led to the analysis of state-of-the-art solutions and tools that will help us develop the AI cyber-defense component that will defend the STAR framework and pilots against poisoning and evasion of AI attacks. Preliminary result shows that poisoning and evasion attacks can significantly affect the prediction performance and reliability of AI/DNN setups, and we highlight the need for deploying defences. A variety of methods have been tested, the design of the AI Cyber Defence tool has been completed and an initial version of the tool has been deployed in the frame of the ongoing integration actions.O6: STAR technologies will start to be integrated and deployed on top of AI platforms at the factories as a set of cross-cutting, overlay functionalities, in-line with the STAR Reference Architecture. O7: STAR surveyed relevant standards and regulations and analysed them from the point of view of STAR developers and users. D2.3 “Review of Applicable Standards and Regulations” has been prepared linking relevant standards and regulations with the scope and activities of the project.O8: The structure, design and the specification of the market platform was finalized and the implementation has started. An initial list of assets, success stories and training resources has been collected to start populating the market platform. 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) STAR aims to advance the state of the art of AI in manufacturing in complementary ways: (i) through the implementation of a pool of leading edge AI systems for simulation and accelerated knowledge acquisition and their customization for popular manufacturing applications; (ii) through the integration and use of advanced AI techniques that render the operation of AI systems more transparent and robust, especially when operating in dynamic environments. STAR will deploy research, develop and demonstrate advanced RL systems, which involve interaction between the human and the AI system as a means of alleviating uncertainty and increasing the robustness of the AI systems operation. Moreover, STAR will integrate novel XAI systems for manufacturing, as a means of increasing their transparency and trustworthiness; (iii) through implementing a set of security and safety functions over different types of AI systems (i.e. deep learning, RL and smart objects) as a means of boosting their trustworthiness and ethical nature.The social impacts foreseen from STAR are summarised as following:1) Ethical AI in Manufacturing: STAR will boost the development, deployment and operation of Ethical AI systems for smart manufacturing based on solutions that boost the security, the trustworthiness, the robustness and the safety of AI systems in digital manufacturing. 2) Worker Safety: STAR will have a positive impact on the safety of factory workers, through providing the means for avoiding hazardous situations that may result from malfunctions or poor operation of robots and other AI systems, especially in cases where humans collaborate with them.3) Increased Workers’ and Manufacturers’ Trust in AI systems: STAR will boost the robustness, safety and trustworthiness of AI systems in manufacturing environments, which will increase stakeholders’ (including workers and manufacturers’) trust in AI systems. This could greatly facilitate the wider deployment and use of AI in manufacturing.4) Compliance to Safety Standards and Regulations: STAR will facilitate manufacturers and providers of AI solutions in complying with safety regulations and standards, which will help stakeholders create safe and compliant environments, while reducing the probability of adverse effects of non-compliance (e.g. regulatory penalties). Related documents Picture1.png Picture2.png
