D2.2 Learning Experience Content Model (draft)
This deliverable proposes a draft specification for the WEKIT content model for learning experiences. The proposed content model consists of an activity modelling language and a workplace modelling language, both together able to express how users of the WEKIT.one AR soft- and wearable hardware solution can interact with media and the environment in pursuit of developing new (or upgrading existing) knowledge, skills, and other abilities.
The core of this deliverable comes in a different format: The specification is a formal submission for the P1589 working group of the IEEE standards association. The new working group p1589 for ‘Augmented Reality Learning Experience Models’ (AR-LEM) was approved by the New Standards Committee (NesCom) of the IEEE standards association on February 16, 2015. The working group was granted until December 2019 and it is under the sponsorship of the learning technologies standards committee and its chair Avron Barr.
The spec draws wide input from the stakeholder community of WEKIT, with contributions formally honoured in the section ‘participants’ in the spec. While the kick off meeting of the working group took place virtually on June 3, 2015, the WP2 work on the content model was organised alongside in the ARLEM working group meetings that took place mostly on a monthly basis. The website for the standards outreach is http://arlem.cct.brookes.ac.uk.
Following this submission, a first ballot will take place, requiring two prior readings in the p1589 working group meetings to establish voting membership. Subsequently, a phase of outreach events shall assist especially technology providers in implementation and validation of their support of the interchange format. The final first version of the standard is planned in for the end of 2017.
The ARLEM specification serves as the interchange format, providing the representation needed to collate recording and sensor data into consumable learning activities and workplace models needed for the re-enactment system for performance augmentation.
D2.1 Functional and modular architecture
This deliverable reports on the software architecture for realising a modular and distributed system for capturing and re-enacting experience. The architecture serves mainly two purposes: building a shared understanding amongst the contributors and guiding the further development by breaking down complexity, thus allowing for distribution of work. Additionally, the structure and its compartmentalisation can be re-used to inform other projects in the context of WEKIT exploitation and beyond.
The proposed architecture breaks complexity down into three layers, the presentation layer, the service layer, and the storage layer (including local and cloud-based data sinks). It documents the components needed to implement each layer: the presentation layer consists of the recorder module, the live guide (re-enactment) module, the analytics module and the community platform. The service layer provides two processing units (for sensor data processing and analytics processing), as well as the following service interfaces: Experience Capturing API (XCAPI), Experience Logging API (xAPI), the Query and Report Interface, Single-Sign-On Services (SSO), as well as non-exposed APIs for sensor data storage and Learning Experience Model storage and retrieval (LEM API). Six data storage solutions are foreseen to hold all relevant data: Sensor Data Store, LEM repository, system logs, a Learning Record Store (for xAPI statements), the user database as well as the community database.
A conclusion and outlook at the end of the document documents the next key milestones for fully implementing this architecture – and for documenting the final version with feedback from the validation trials.
D3.3 Software Prototype with Sensor Fusion API Specification and Usage Description
This deliverable reports on the components of the first functional prototype of the WEKIT sensor fusion API and the experiences made with it. The development of these components is based on previous work on the WEKIT framework and methodology, requirements and scenarios, as well as technological selections and limitations. The deliverable specifies the key interfaces between the software component and the hardware, the backend infrastructure, and the front-end application modules. To create these interfaces, a corresponding data structure that allows to store the learning materials is defined. This learning materials include:
- A data structure to define the actions performed by a learner or expert to complete a task
- The captured learning experiences
- Annotations to the actions
Following this data structure the document also provides a proposed interface to store, retrieve and manipulate the learning material.
Furthermore, the deliverable contains usage recommendations.
D3.2 Hardware prototype with component specification and usage description
In this deliverable we present the first version of the experience capturing hardware prototype design and API architecture, following on from D3.1 and the final selection of sensors. This deliverable involved acquiring, testing and integrating the various off-the-shelf sensors and developing a hardware/software design to connect the various devices and sensors into a single platform. This also involved solving problems around how we would manage the additional computing power, storage and wireless streaming capabilities required for this project.
In this first version of the deliverable we propose an initial design for the hardware and architecture and have focused initially on providing a bare-bones prototype of the Sensor Processing Unit which involved combining a micro-pc with various sensors to test the flow of data, capabilities of the system and possible connection methods to the device for the various sensors.
The first hardware prototype does not yet include the wearable element, however, our selection of the sensors and micro-pc takes into consideration the desire to integrate the hardware using 3D printing or by creating add-ons for the Hololens to avoid a chunky or multi-device hardware solution. We have provided a mock-up in the Prototype and Usage Descriptions section below of what the final prototype is expected to look like.
As described in D3.2 although there are other AR alternatives already available in the market, they all fall short in terms of the required spec and functionality. A decision was then made to work with the Hololens as it would be the best in class and although not available has been tested by key partners and will be available in time for the next deliverable.
Within this deliverable we also provide an initial mock-up of the hardware design and also a sample of the API and links to device APIs.
D7.2 Outreach & Dissemination Report 1
In this first iteration of the Outreach and Dissemination Report – D7.2, we report on initial efforts and outcomes in dissemination and stakeholder outreach for the first year of the project. We thereby adhere to the planned structure of the original dissemination plan in D7.1. As planned for this early phase, we do not yet have shareable and tangible project results, but there is much external interest in our framework and vision; the consortium has not only reached but exceeded its target measures for five milestones (relevant stakeholders identified, community portal and social media online, community event planned, prepared, and organized, first stage of 30 members on community portal). Lists of relevant stakeholders and relevant stakeholder outreach events, particularly WEKIT community events, are continuously filled with information from all partners as online internal collaborative resources, also used as internal reporting instruments. Partners have disseminated informational and promotional contents on WEKIT on a wide basis with the help of online multimedia (video trailer with over 700 watches, blog articles, newsletters, online journals), print media, public presentations on conferences and fairs, as well as face-to-face meetings with policy makers. In particular, the consortium held the first WEKIT Community Event at the European Conference on Technology Enhanced Learning in Lyon, France on 13–16 September 2016, thus reaching an audience from the wider Technology Enhanced Learning (TEL) industrial and academic community. Altogether, including WEKIT appearances at over 20 public events, some of them with participation of key innovators and policy makers in project related areas, the WEKIT message reached more than 56.000 people. These events and the accompanying messages in social media and our community portal produced correlated peaks in the different web-based social media channels, which in this first project phase produced moderate international traffic, mostly from Northern Europe and the US. The WEKIT website registered over 250.000 hits in more than 15.000 visits from more than 9.000 unique visitors.
This deliverable contains adjustments of the initial dissemination plan for specific activities in the second year of the project and some general updates on the dissemination and outreach strategy.
D6.3 Training Scenario and Evaluation Plan for Space
This deliverable describes the WEKIT evaluation case for Space. The space domain is a very complex and challenging environment where the range of pre-flight and in-flight activities that can benefit from the innovative and powerful forms of AR-based and knowledge-intensive training that WEKIT is aiming to develop, verify and make available to advanced industries across Europe.
Prior to WEKIT, a variety of activities leads not only to the production of huge amounts of data whose significance can be hard for trainers to interpret, but also to the timely development of instances of personal know-how that are hard to capture, share and deploy. Our space pilot and evaluation explores more effective ways for trainers and operational managers to handle that flood of data and to transfer and exploit knowledge among the actors involved in a project or a mission.
In WEKIT, we define use cases that test the scope, power and usability of our methodology. These scenarios are described in order to create solid cases to be tested in the working environment to prove that the WEKIT prototype can enhance the performance of the activities in such a field. In particular, the purpose is to reduce the numbers of mistakes during the assembly, maintenance and training activities, reduce the time needed for training (including training in how to speedily detect and recover from mistakes) and be the basis for a new concept of training where the trainer and the trainee can be remotely connected from different locations. Success in that range of space-industry trials is highly likely to result in significant advances in Europe’s capacity to rapidly master innovations in a wide range of knowledge-intensive industries and contexts, with the implication that this deliverable will have high strategic value for years.
D6.2 Training scenario and Evaluation Plan for Engineering
This deliverable describes the WEKIT evaluation case for Engineering in healthcare. We identify scenarios and examples of use cases to demonstrate how the learning processes may be improved by using the WEKIT methodology, in terms of effectiveness, time reduction and user perception.
the medical domain is a domain in which complex learning occurs. It involves understanding heterogeneous physiological systems, developing adaptive expertise and acquiring the collaborative skills required in multidisciplinary medical practice. It involves mastery of competencies that enable the individual to effectively perform occupational activities to the standards expected in the professional environment. This requires ample opportunity to practice and the ability to experience all possible variations in contexts and circumstances in order to reach the expert level. Healthcare learning scenarios provide a big variety of use cases as well of different end-users that could benefit from the learning experience.
By way of example, the pulmonary embolism disease has been chosen as a specific clinical use and a protocol for achieving a correct diagnosis is presented by steps. The presented WEKIT training approach, anyway, can be generalised and applied also to the analysis of different diseases and pathologies.
D6.1 Training scenario and Evaluation Plan for Aeronautics
This deliverable describes the WEKIT evaluation case for Aeronautics. Three tasks of different complexity are presented: engine rigging, brake wear check, and preflight inspection.
1. Complex maintenance task. Aircraft engine rigging involves adjusting various components associated with the control systems in order to get a smooth performance of the aircraft during ground running and flight, and the associated parameters within limit. It requires strict conformance to procedures described in the manufacturer`s maintenance manuals and service instructions. For this, technology can be used to provide better training, sharing of experience, and capturing best practices. Such a solution can save time and reduce the associated costs.
2. Non complex maintenance task. A brake wear check is performed at preflight inspections and every 200 hrs phase inspections. Inspections are performed to measure the brake wear. On this task and others, the maintenance manual pictures are not always easy to understand. For this augmented reality glasses can be used for object enrichment for example to display arrows were to measure.
3. Medium complex maintenance task. Pre-flight inspection is used to determine if the aircraft is in airworthy condition. In order to conduct a pre-flight inspection, a lot of paperwork and reference information is gathered and studied before actually proceeding to the aircraft to conduct the inspection. To this end, an automated solution can reduce the time and improve the quality of inspection.
D6.0 Training Scenario and Evaluation Plan Methodology
This deliverable has the objective to introduce the work done within WP6, explaining the innovative methodology used to define and evaluate the Industrial cases.
Our starting point is the need to evaluate the learning methodology and the technological platform in authentic contexts. The project will implement a prototype and conduct two cycles of trials in actual workplace settings. Based on the WEKIT methodology, a set of scenarios will be developed for end-user-prioritised tasks in three distinct industries (aeronautics, engineering and space). The scenarios will be described in D6.1, D6.2 and D6.3 while their evaluation will be described in D6.4, D6.5 and D6.6. This provides the basis for developing analytics for the impact and effectiveness of the industrial learning methodology.
We will facilitate the cooperation among the Industrial Partners and ensure the consistency between the pilot cases and the WEKIT methodology. We will collect the requirements from the Industrial Partner and from the WEKIT community in order guide the development of the prototype and to create solid cases to allow a thorough evaluation of the prototype.
During the trials both quantitative physiological and performance data will be collected in order to get a robust triangulation of the results and feedback collected during and after the trials. The results of this test and evaluation phase will be used to define a roadmap for a full exploitation of the results.
D3.1 Requirement analysis and sensor specifications – First version
In this first version of the deliverable, we make the following contributions: to design the WEKIT capturing platform and the associated experience capturing API, we use a methodology for system engineering that is relevant for different domains such as: aviation, space, and medical. Furthermore, in the methodology, we explore the system engineering process and how it can be used in the project to support the different work packages and more importantly the different deliverables that will follow the current.
Next, we provide a mapping of high level functions or tasks (associated with experience transfer from expert to trainee) to low level functions such as: gaze, voice, video, body posture, hand gestures, bio-signals, fatigue levels, and location of the user in the environment. In addition, we link the low level functions to their associated sensors. Moreover, we provide a brief overview of the state-of-the-art sensors in terms of their technical specifications, possible limitations, standards, and platforms.
We outline a set of recommendations pertaining to the sensors that are most relevant for the WEKIT project taking into consideration the environmental, technical and human factors described in other deliverables.
Finally, we highlight common issues associated with the use of different sensors. We consider that the set of recommendations can be useful for the design and integration of the WEKIT capturing platform and the WEKIT experience capturing API to expedite the time required to select the combination of sensors which will be used in the first prototype.
D1.1 & D1.2 Training Industry Needs & Technology Industry Needs
This deliverable joins D1.1 (User Industry Needs) and D1.2 (Technology Industry Needs and Affordances) and reports on the outcomes of Tasks T1.1 (Training Industry Assessment) and T1.2 (Technology Industry Assessment).
This deliverable gathers information about current practices in the training industry about the current and potential use of AR/WT in educational processes in order to assess training industry needs (D1.1) and to extract input for the WEKIT Framework and Training Methodology (T1.1).
It also gathers information about current practices in the technology industry about conditions, success factors and acceptance conditions for the current and potential future use of AR/WT in educational processes and other industry relevant use cases in order to assess technology industry needs (D1.2) and to extract requirements for WEKIT scenarios and technical prototype (T1.2).
D1.3 WEKIT Framework & Training Methodology – First version
When mapping the transfer of a skill or some knowledge between and expert and a novice there are many high-level tangible and intangible factors that are present. Three key factors are the type of task to be carried out, what domain will be interacted with and dimensions that define the tasks constraints.
Tasks are activities that work with the perceptual, cognitive and developmental aspects of the person being trained. The Domain is what is being interacted with that might be a human being (the trainee or their colleagues), a machine (some equipment) or a specific environment (for example, a vehicle’s internal space or external atmosphere).Lastly there are Dimensions that are the boundaries for the task that specify the space it will be carried out in, time factors that relate to how the timing of the task unfolds as well as the key aim, target or intention for the task.
When a use case, comprising of specific activities to transfer learning, is mapped, there will be specific attributes that will be assessed which will relate to one or more combinations of Task, Domain or Dimension. When specific types of Task, Domain and Dimension are assigned weights of importance to an attribute, the scores can be added up to identify which factors are the most critical for an attribute, the learning activity as well as the use case as a whole.
These patterns will be overlapped with the pattern of a bottom up level task class which will identify dominant skills pattern patterns required to perform the job. This pattern of scores can be used to define the requirements for training scenarios that will finally be captured, enacted and assessed. Then proper transfer mechanism may be then selected to design the learning platform which hosts a set of requirements for capture and enactment of the skills.
D1.4 Requirements for Scenarios and Prototypes
In WP1 the WEKIT consortium develops a framework for wearable experience, specifies a corresponding methodology for vocational training, creates suitable application scenarios, and derives requirements for the technological platform accordingly. The first findings are documented in the Deliverables 1.1-4. This deliverable (D1.4) is the first outcome of the WEKIT Task 1.4 Requirements for Scenarios and Technological platform, where the stakeholders can continuously collect, update, and negotiate requirements for Wearable Technology (WT) and Augmented Reality (AR) solutions, which should be developed in this project. The end-user requirements have been elicited through co-design activities and captured in Requirements Bazaar, a social requirements engineering toolkit that was initially developed by RWTH in the ROLE project and was awarded the best demo paper award at the IEEE International Conference on Requirements Engineering in 2013. Later on it was successfully used also in other projects, including Learning Layers. The House of Quality (HoQ) approach was adopted to offer a consulting tool for assessing technological options and comparing available tools for AR based workplace learning support. Here we present the current status of the collected requirements, which will inform our developments in the technical work packages WP2-WP5. This report will be later on updated in M21 and M36.
D7.1 Dissemination Plan
The WEKIT dissemination plan provides a structured planning of relevant activities within the timeline of the project. The plan contains the dissemination aims and principles, the target audiences, the WEKIT communication strategy, the communication actions and tools, communication management plans, as well as the measures of success.
This document focuses on a specific yet important aspect of the project’s dissemination and communication strategy: identifying the target groups that need to be reached as well as a timeline for dissemination activities.
Certain issues and quality criteria should be considered during the implementation of dissemination actions. This deliverable also tries to give useful recommendations on this. Furthermore, some instruments to evaluate the dissemination activities will be presented.
A project and its results can only be successful when they are of sustainable value, both during and after the project timeline. The actual use of project outcomes is dependent upon successful dissemination activities. Therefore, it is important to understand that a good dissemination strategy is the basis for successful exploitation of results.
D8.1 Exploitation and IPR Plan
This document outlines our evolving framework for managing results and linked IPR from WEKIT and similar projects. Our ‘work-in-progress’ approach is presented as a storyline based on a learning journey. The journey begins with desk research into current good practice relating to results and IPR, and ends with examples of incorporating innovations such as (WEKIT-linked) evolving open standards and workflows that are being developed by industry and standards groups. As with all learning journeys being developed in WEKIT, each step in that self-referential journey could be replicated, shared and experienced using WEKIT tools. The purpose of such an experience would be to appreciate and then to master results-focused product and process innovations covering every step in managing results, starting with identifying problems and challenges that lack an effective/affordable/scalable solution. Seeking a solution is the driver for investing resources in obtaining a result that can lead to a faster/better/cheaper solution. The learning journey for managing results then looks at how to focus our resources on driving for high quality results. Quality here means: making a significant difference to the state of the art; easy to appropriate (adopt) by a wide range of beneficiaries; can be the basis for self-sustainability after a project’s initial funding ends.