CREATION OF A COLLABORATIVE ROBOTICS’ LOGISTICS STATION: Workers assistance in assembly, disassembly and checking parts: The Sharework project implemented Cembre’s vision of a new manufacturing shop floor by introducing collaborative robotics technologies to improve working conditions and increase productivity.

The main changes to be implemented in Sharework project were to create a logistics station where collaborative robots and humans work together to check the pieces and load and unload them on metallic plates.

Sharework provided a flexible transportation system connecting the assembly stations with the different machines and integrated robots into the workshop for assisting operators when moving across the station, understanding operators’ behavior and anticipating tasks. Robots supported operators in repetitive assembly tasks, improving their ergonomics.

updatereview of KPIs that will be used in the final demonstrators

Provides a multi-layered security defence from potential external malware intrusions and threads. The security module communicates with multiple detection systems, triggering appropriate responses in the physical and digital world if needed.

_____

Data Security or Cyber Security is a subject that becomes more and more important as cyber-attacks increase in both frequency and intensity. This document presents the related findings and developments of the ShareWork Task T5.4 “Data Security”. This document reports on the policies and the technologies selected to ensure data security on the system in terms of privacy, identity and integrity. The importance of the Cyber Security is presented through an analysis of security threats for the SHAREWORK cyber physical, Human Robot Collaborative system. SHAREWORK. Cyber Security threats in HRC systems are especially important as their results can range from data theft, production, equipment or product damage to human injuries or deaths. While the analysis of the threats and impact are focused on the ShareWork, the authors tried to generalize the findings so the analysis of potential threats and impact could be also used as a reference.

In addition, a comprehensive and extensive set of Security Policies is provided following and implementing related sections of recognized Security policies, such as the NIST Security Framework, the CISA Cybersecurity & Infrastructure Security Agency and the CIS Critical Security controls. The SHAREWORK security policies provide details on how to implement the prevent, detect and respond functions of security. In particular policies, technologies and measures are provided on how to protect the system in different levels of the architecture.

The ShareWork data security prevention function policies take into consideration the ShareWork architecture to propose technologies, and administrative measures and policies for security in the software, hardware, physical and network level. Afterwards, the Data Security module is presented, demonstrating the importance of the detection and response functions in a human robot collaborative environment.

The SHAREWORK prevention data security policies provide multilayered security following a defense in depth approach. In addition, the Data Security module integrates and connects with detection systems in different layers to identify potential threats and trigger appropriate responses in the physical and digital world. In particular it triggers other ShareWork modules and can even connect with hardware such as a PLC to shutdown critical operations. Furthermore, the Data Security module informs different human stakeholders using audiovisual means. Email and advanced integration with Module 9 “Direct and natural human system and system-human communication” that enable immediate operator notification. In addition, the potential threats are logged and can be viewed to identify patterns or trends in data security incidents.

Multi-modal interface for Human-Robot and Robot-Human interaction working in three interfaces: an Augmented Reality application, an Android Tablet application and a Smartwatch application.

Recognizes if specific bolts have been inserted into the componing. It achieves this by analysing the output of the ZED camera mounted on the robotic arm to frame the entire surface of the component. The application is built in such a way that it can be integrated and adapted to similar use-cases with low effort.

Continuous evaluation of human ergonomics and taking into account human-factors within the development phase.

Enables or switches off safety measures and sensors on the basis of the level of cooperation among the human and the robot.

Recognizes the task the human is performing. This allows to trigger the “Human-Aware dynamic planning and scheduling for HRC tasks” module and plan the next action that the robot should perform according to the human actions.

Tracks in real time the position of the operator, to identify the corner where the operator is working, and the body posture of the worker, to avoid collisions.

Triggers and monitors the execution of a set of tasks defined along with their temporal execution. The task planner must be constantly communicating with the knowledge base in order to be updated with the actual state of the system.

Represents the information flow of the entire system. It contains updated information about the use-case, the task-diagram flow, the location of each task in the use-case, and whether the task must be executed by the worker or the robot.

Plans a path from the robot current position to the received goal and executes the trajectory while taking into account the presence of the human in the workspace.

The video features the prototype at STIIMA-CNR

Augmented reality application, hosted on HoloLens, that serves as a training framework for novice operators during their acclimatization with collaborative robotics.

Analysis of the human movements for a possible robot motion planning. As input the real time skeleton of the human detected by the Human Tracking Module and depending on the position of the joints of the skeleton returns as output the recognized gesture.

It provides the robots reactive and efficient decision-making capabilities to adapt to highly unstructured environments, as the ones in which the human operator might constantly enter in and modify the workspace. In a situation in which an operator enters in the workspace, the robot should be able to efficiently recompute its future actions to safely adapt to the human. To do so, a set of algorithms have been developed for human-aware task and motion planning. To safely adapt the robot motions to humans, dynamically adaptable, human-aware, motion planning algorithm has been developed, in collaboration with CNR. The algorithm has two main elements. On one hand, a human occupancy prediction algorithm has been developed. Given the current human state, the algorithm predicts the probability density function on where the human will be in the future. Then, the motion planning algorithm, developed by CNR, takes this information into consideration to plan safe motions in the workspace. The problem associated with modelling and learning the human occupancy prediction modules has been addressed. Based on highly expressive probability density function models, known as Normalizing Flows; in our work, we explored the integration of these models for the problem of learning conditioned human occupancy predicted densities. Additionally, we have explored the possibility of using previous data to guide the search in motion planning and improve its efficiency. By using CNN and ANN as approximators, we trained predictors on the feasibility of some key decisions in robot grasping and further derived heuristics from those predictors to accelerate motion planning. 

Provides a real-time probability of the human state according to a learned distribution to refine the path search based on the current state of the human and the context.