Category:Training

From EuroVR Knowledge Base

Contents 1 Overview of application area 2 VR training applications used by industry 3 Examples of potential scenarios 3.1 Training in the workplace 3.2 VR training authoring tool 3.3 Remote training and continuous education 3.4 Training and task planning in hazardous areas 3.5 Training for interacting with robotic systems 4 References

Overview of application area

Virtual Reality (VR) technology is perceived as a powerful tool for training applications as it enables convincing virtual environments (VEs) which are an important characteristic for simulating real-life tasks. Training through VR technology is recommended in the following situations:

• where real-life operations are too dangerous
• where real-life operations are too expensive
• where real-life operations are not possible
• where real-life operations are complex to learn

The main benefits of using VR over traditional training methods are listed below:

• Reduction of training costs
• Increase in safety
• Reduction of downtime

Training through VEs is based on the idea that the user experiences better learning and retention of the acquired knowledge with practice. Simulated scenarios help the user learn faster and remember procedures better. These scenarios enable users to put into practice what they have learnt, and confront them with the consequences of their decisions. In a training process, the main goal is to acquire or improve knowledge and skills for the task's performance by the provision of information or experience. According to “power law of practice” (Farmer et al., 1999), the time required to perform a task is directly proportional to the performance time on the first trial and inversely proportional to the trial number. It is clearly observed that the process of task repetition will achieve better results and this is not a problem in a virtual environment because it can be repeated indefinitely. In the context of training, Farmer et al. (1999) talks about the main skills associated with a user and about the kind of instructional activities. They classify skills in this way:

• Perceptual-motor skills: This involves co-ordination between stimulus or perceptual input received from the environment, and the motor responses to the received inputs.
• Procedural skills: This involves the execution of an algorithmic sequence of discrete actions for the performed task. An important issue is the determination of the kind of information incorporated within the mental model, how this information can be acquired most efficiently and how it affects transfer and retention.
• Cognitive skills: This is the synthesis and integration of different types of information which have been acquired through human senses and made understandable to the user.

The natural way of performing the tasks influences the perceptual-motor skills, whereas intuition doesn’t influence the procedural and cognitive skills.

Furthermore, instructional activities can be distinguished in terms of whether they occur prior to, during or after training sessions. These activities can be labelled briefing, tutoring and debriefing. The major distinction between tutoring and briefing/debriefing is that tutoring is a process in real-time whereas briefing/debriefing is a process that proceeds off-line. Furthermore, the type of interactions (navigation/selection/manipulation) with the objects which compose the virtual training world can influence the training sessions.

Navigation is a basic mechanism of interaction that is required in many immersive virtual applications and based on the control of the user’s viewpoint motion in the three-dimensional environment. In most cases, navigation is not the specific goal. Rather, it is simply used to move the user into a position where he can perform some other, more important tasks. Because of this, the navigation technique should be easy to use, easy to learn and cognitively simple (the mechanism of navigation should be transparent for the user).

The selection process involves the choice of one or more virtual objects by the user for some purpose. Often, selection is performed to set up manipulation, which sets the position and/or orientation of a virtual object. Obviously, unless the user is constantly manipulating a single object, he must first select the object he wants to manipulate. Some virtual objects can be reached directly by the user (user’s hand maps to the virtual hand) and for some others, the user must extend his virtual hand much further than his physical hand.

In order to perform useful training, the interaction mechanisms must have the following characteristics:

• Transparency for the user.
• Ease of learning.
• Efficiency in the task performed (more accuracy and velocity).
• Comfort in simulations of large periods of time.

VR training applications used by industry

• FIACRE An application to teach the train driver to test and handle switch blades on high speed tracks.
• SIMURAT An application to teach the freight agents of the rail company to detect defaults in wagon infrastructure and superstructure before letting them roll on the tracks.
• INSCAPE (2004-2008)  (INteractive Storytelling for Creative PEople) Development of an authoring platform that aim at enabling non computer scientits person to use and master the latest Information society Technologies to build interactive non linear stories and applications
  
• CRIMSON Crisis Management Solution - The CRIMSON system combines the latest Simulation and Virtual Reality technologies for the inter-organisational preparation and rehearsal of security missions in response to major crisis (natural events, industrial accidents, NBRC incidents...).
  
• WAVE CS WAVE is a virtual welding environment dedicated to welding training. It does not replace existing training. It brings a pedagogical improvement through the optimisation of the gesture learning and the concentration requested from the forthcoming welders.
  

Examples of potential scenarios

As part of INTUITION the working group on education and training discussed how VR could be used in the future for training. The following are scenarios as a result of these discussions.

Training in the workplace

Bob is a maintenance engineer. When he arrives at work he is automatically recognised as he enters his office and the computer tells him his work schedule for the day. One of his tasks is to repair a machine and he is alerted to the fact that his training for this machine needs to be reviewed. He selects the link and this shows him his training profile. He can see that the health and safety regulations have been changed since he last used this machine and he must update his training. An hour slot has been included in his schedule to allow for this.

He chooses to carry out the training at work in the virtual training centre. He sometimes prefers this as training from home or his office is sometimes hindered by distractions, e.g. the phone, the kettle or people generally moving around. Besides, the training centre often has the latest technologies and devices which are not yet affordable to the domestic market. Also he does not have the space at home for a full CAVE™ system.

At the allotted time he enters the training centre, which is essentially a 6-sided immersive room. He is greeted by the computer which automatically recognises him and the training he requires. Bob requests a virtual trainer embedded in the system. The virtual trainer is an avatar of a maintenance engineer similar to Bob. Over time he has been adapted to match Bob’s needs and identify the most appropriate learning style for effective training in the minimum amount of time. However, a variety of methods of training are always offered - in case Bob would like to try something different today to make the training more interesting.

The variety of methods includes: 1. Reading the health and safety regulations documents to find out what changes have been made 2. Watching a video that shows how these changes affect use of the machine 3. Watching the virtual trainer working through a task 4. Requesting a real trainer to work through the task 5. Requesting to view a colleague’s (with permission) recorded actions of the task 6. Virtual hands-on experience of the task o With guided instruction (text, visual prompting, voice-over, virtual trainer, real trainer) o Without guided instruction o Choice of a variety of displays (projected, headset, holographic) o Choice of a variety of devices (devices to control virtual tools, tangible tools)

Bob decides to ask his virtual trainer to choose. He recommends that Bob watches him first complete the task and then tries it himself. Bob views the task being performed on the holographic image of the machine in the centre of the room by his virtual trainer. Bob is able to walk around and view the procedure from different angles. When he feels ready, he picks up a real automatic screwdriver which is an interaction device linked to the VE and he attempts to carry out the task. The tool sounds real and provides realistic feedback, and this, combined with visual confirmation of its actions, makes Bob feel like he is actually performing the task on the real machine.

Bob completes the task a number of times and views these on the replay mode to improve his performance. Finally he is ready for evaluation. He carries out the task and then receives feedback. He has achieved his health and safety certification. The virtual trainer updates the status of his training profile and recommends other training courses. Bob decides that he would like to do some more training and tells the computer to send him a message in a month’s time to look into the new courses available.

VR training authoring tool

Wendy is one of the company’s remote trainers. The company has a number of these distributed workers which means that they do not have the overheads of a large office and can be responsive to their customers located worldwide. The company have just delivered their new product to their central warehouse. Before leaving home Wendy has downloaded all the specifications for this product on her PDA. Her job is to design the training programme. Using the specifications on her PDA, she can connect this to her glasses which enable her to receive information about this product. This mini augmented display enables her to strip away layers of the product and view the internal mechanisms. She can also receive further information about each of these parts and/or watch 3D animations explaining the key features of the product, how it works and how the customer can use it or interact with it; during animations, she can also change the time-scale or move along the time axis, and she can interactively move her point of view. . Using the VR training authoring tool, she develops an appropriate VE adapted from reusable models from previous training programmes. This new programme can then be downloaded by their clients when they receive shipment of the new product.

Remote training and continuous education

Travis is an engineer. The company has recently bought another organisation located in Germany. He decided that it would be beneficial and interesting to get a qualification in Engineering German. The course is accessible from home. It has a variety of options for learning, ranging from individual learning, networked with other learners and lecturers, to opportunities for summer schools etc. For Travis, the greatest advantage of using virtual environment learning is the opportunity to practice the language in some virtual context. Travis can enter a virtual German factory and discuss topics which he has just learned with a virtual or real German engineer. He is able to request instant verbal feedback as he speaks, replay the scene to view his performance and view the conversation in text to support his written learning. He can also practice the scene a number of times until he is satisfied with his progress. Travis is able to engage with virtual German engineers from different regions to test his ability to recognise the language with regional variations, and request different levels of sophistication from basic engineering language to complex concepts.

He also has access to real lectures via the internet. All he needs to do is check on the schedule when a lecture is likely to take place in a German institution or in the UK, and book a slot. He has the opportunity to either monitor the lecture or choose the option to participate. This means that he is able to interact in the classroom by answering or asking questions.

Training and task planning in hazardous areas

During normal operations and principally, during operations in outages of Nuclear Power Plants, there are some activities that must be carried out in hazardous areas not frequently accessed. The virtual representation of these areas should make it possible to point out some relevant safety aspects such as: • Radiological map and Hot Spots for training, allowing user to classify areas depending on the potential risk (also useful for re-designed areas according to Radiological Protection requirements) • Calculation of distances and real measurements of hazardous areas in order to plan future tasks inside. Some maintenance activities, even if they meet the ALARA criteria (As Low as Reasonably Achievable, applied to committed radioactive doses) can not be carried out due to space limitations, (e.g. equipment movements, maintenance operations using heavy machinery, scaffolding, etc.) In addition, it would be necessary to be able to interact with the virtual environment for: • Checking if a piece of equipment could be sited in a given area • Creating routes to identify the most appropriate one (both for training and tasks planning)

The operational conditions of these hazardous areas impose the need to protect the workers, minimizing the time inside. In other cases, some maintenance or repairs tasks need to use machinery with determined measures, and it’s necessary to check if they could access these areas and could fit into the required place.

Training for interacting with robotic systems

Philip is an astronaut; his next mission will consist of an extra-vehicular activity to repair a damaged module of the International Space Station. He will receive the support of a robotic agent which is able to perform a number of complex tasks in microgravity conditions. Philip needs to train himself to correctly interact with the robotic agent in order to achieve a successful result during the real mission. The robot is equipped with some basic form of artificial intelligence, meaning that it can react to some specific actions or requests required by the human operator. Philip has to learn how to correctly and safely interface himself with the robot and how to synchronize his actions with the robot, i.e. how to collaborate with it; in particular his main goal is to learn about possible problems related to safety and time constraints when collaborating with the robotic agent. In the Virtual Reality training centre, Philip can simulate the repairs mission in a Virtual Environment where a virtual model of the robot is acting; the virtual robot is provided with: - a behavioural model which faithfully corresponds to the one of the real robot; - a realistic dynamic model to simulate in real-time the physical behaviour and capabilities of the real robot (i.e. to detect collisions, to handle virtual objects, to take into account the effect of microgravity, to simulate the realistic motion of the robot itself and its mechanical arms, etc.) Also, Philip can receive haptic feedback from the simulation when he manipulates objects. The level of realism of this VR application is mainly focused on the physics simulation and appropriate haptic feedback. The visual realism is also high, but it is not a crucial aspect of the simulation.

References

• Farmer, E., Van Rooij, J., Riemersma, J., Jorna, P., and Moraal, J. (1999). Handbook of Simulator-Based Training. Ashgate Pub. Ltd., Hants.