How we develop a VR learning environment

1 Sep, 2016

“How do you go about developing a VR environment – and what do you need to know about my business first?”

When working with new clients, we’re often asked about how we build VR learning environments. While customers understand that VR can be used for training everyone from submariners to surgeons, they’re not always clear on the processes involved or the inputs required.


What can a VR learning environment do?

Using a VR learning environment is like being inside a custom video game, where the tasks and equipment are replicas of those in a real workplace environment. Submariners, pilots and tank drivers use VR environments to learn the controls of their vehicles. Surgeons use VR environments to learn theatre techniques without posing a harm to real patients. Mechanics use VR environments to explore new engine technologies and observe machine processes close up.


What do you need from me to start developing my VR environment?

Firstly, we need to understand your business context to help guide you towards the right solution for your timescales and budget. We’ll begin by asking key questions such as:

  • What are the goals and objectives for your training programme?
  • Who are the target audience? How many will undertake the training each year?
  • What are you training people to do?
  • Does your VR course sit within a wider training programme? If so, where?
  • Where will learners complete the training?
  • Is training self-directed or trainer-led?
  • Did the training exist in a previous form? If so, why is it being replaced and what metrics will demonstrate improvement?
  • What are the timescales for the project? Have milestones already been set?
  • What are the success criteria?

These may all sound obvious, but it’s surprising how easily this type of vital information can be overlooked or assumed.

Secondly, we’ll need understanding of the training itself. Often, clients already have a strong view of what should be included in their VR experience, and how it should look and behave. Our initial approach is to challenge these assumptions in order to ensure we’re not wasting time or budget. Why build an entire production facility if 90% of the training will take place on one specific piece of machinery? The problem and the desired outcomes must drive the training experience, not vice versa.

Answering the questions below will usually tease out the detail we’ll need in order to get started:

  • What virtual 3D environment will the training take place in?
  • What kind of objects will the environment contain?
  • How accurately do the objects and the items have to be rendered?
  • What kind of reference materials are available? Are there existing models? Videos? Photos? Diagrams?
  • What key actions are learners expected to take?
  • Roughly how many action steps are required for successful completion of a task?
  • What types of reactions occur when correct or incorrect actions are made?
  • What level of simulation is required? For example, are actions linear or non-linear? Does pushing button A have one result? Or does it have many depending on my previous actions?
  • What metrics do you want to capture?

Once we have this information we can build a Backlog and start developing your training. We’ll almost certainly discover knowledge gaps along the way and we’ll probably ask you questions you don’t necessarily have answers to – all part of developing your effective training solution.


How do you create the 3D models?

Everything learners interact with in the environment is a 3D model; a computer rendering of a real object or place. VR learners can view these models extremely close up and from any angle, so they need to be comprehensive and seamless to ensure immersion. Building these models makes up a high proportion of the work and investment in creating a VR learning environment.

Generic models of factories, oil rigs, offices and so on do exist, but if you need to model your specific work environment (which is best for your learners – learning to work in the actual space they’ll be occupying on the job), we have to work from scratch. If you have existing models from another VR learning environment, we may be able to repurpose those, but mostly we’ll need information about the places and objects from you.

The best thing for us, if there isn’t a model available already, is to visit your site and see first-hand the space and the objects we’ll be reproducing. This gives us a sense of how things connect and interact, and how it should feel to occupy and move around the space you want us to simulate.

The next best thing is a video: not a marketing video, but a walkthrough of the space, a series of images which we can use as the basis for our model. We prefer video to a series of photos because video allows us to see things in action – it’s easier to accurately simulate an environment if we’ve seen it move.

We prefer not to work with technical specifications or manuals, other than for visual reference. They’re generally too abstract for our needs and don’t give a sense of how an environment works. For example, a manual might identify every part of a car engine, but it won’t show us where the fuel goes or how parts move when they’re running.


How do you develop the training?

We’ll essentially run three streams, sometimes in parallel and sometimes in sequence:

  • Art – 3D design, modelling and lighting
  • Game design and development – adding interactivity and logic
  • Backend / frontend development – varied, but can include customisation of non-3D page elements, configuration of reporting, white-labelling

Following Agile Sprint processes, we’ll deliver meaningful chunks of work every two weeks and invite you to take a look. Things may be a little abstract in the very beginning, but you can usually walk around an early version of your environment in the first couple of weeks.

It’s worth noting that we’ll all discover things along the way, and you may change your mind about some facets of the training once you start seeing it for real. All development works this way and VR is no exception. As a very new medium, these discoveries can often be surprising and unlock new opportunities not previously considered. Whatever happens, our processes and your input will ensure this drives us to the best outcome.


To find out more about VR learning environments and the training programmes using them, visit