Australian Mining Simulator

I recently came across a press release from the University of Adelaide in Australia titled “Virtual world is safest for miners.” In cooperation with the University of New South Wales, they are working on creating a simulation that allows miners to train for dangerous work using what they refer to as virtual reality and gaming technology. The usefulness of doing this kind of training seems pretty obvious.

I’ve gone back to school and I’m studying computer science. Since I’ve always been interested in virtual reality, it occurred to me that I would like to do my undergrad and graduate research in virtual worlds like Second Life. One of the first things that occurred to me was the possibilities virtual worlds hold for doing different kinds of simulation and training. One idea that came to mind was the ability to train firefighters in a virtual world. You could perhaps have your crew logged in to SL, all equipped with voice, and they could move their avatars around to put out a simulated fire. It’s interesting to think you could test the firefighters by throwing unexpected situations at them to see how they react. Maybe during the fire you have a floor suddenly and unexpectedly collapse and test how well they are able to cope with the situation. This sort of thing would be ideal for training NCOs and officers in the military to lead troops in different situations.

There are a lot of benefits to this kind of training. The most obvious is safety. Another is the fact that you could burn the same building down several times until the trainees get it right. You can’t do that very easily in real life. The costs would be much lower and there is no harm to the environment.

Another idea I’ve considered is using virtual worlds as a much fuller way to debrief people or teach them about events that have occurred in order to learn from them. I think of the big shootout police in California had with armor-clad bank robbers a few years ago. Imagine if you could recreate that situation in a virtual world. Instead of reading about the event or being lectured on it, or using various charts or slides to explain what the actors in the situation did, imagine if you could show them in 3D space? Imagine the benefits of seeing what it looked like from ground level as it happened. Imagine how useful it would be to stop the action and camera around to show what’s going on from different angles. It would be a much more immersive way of teaching what happened in that situation. As a learning tool you could even stop the scene, rewind it, and then show a better way it could have been handled.

I think we are all familiar with the 3D simulations that many TV shows employ to give us a better view of what happened. But those simulations are not interactive, and they tend to only be used for short periods. It is probably too difficult and costly to put together an entire simulation from scratch. But in a virtual world, the platform is there already. You don’t need to recreate it for each new simulation. And there is the advantage that you could actually bring avatars into the action where appropriate.

One thing that disappoints me about the article on the mining simulator is that they refer to “gaming” technology. I think that could tend to hurt the credibility of the project. I think I understand why they put it that way, but it tends to make one think that the simulation is some sort of first person shooter. One problem with the first person shooter idea is that you can’t simulate the physical actions of mining through a computer interface. If you play sports games on a console, you may feel immersed in the game, but the simple fact is that pressing the “X” button in a baseball game teaches you nothing about hitting a slider. There is no physical correlation between the two things. But Dr van den Hengel, one of the heads of the project at Adelaide says something very interesting about the simulation:

“Virtual reality simulation and computer gaming are powerful tools for conditioning human behaviour,” he says. “Both technologies enable users to experience a range of situations that would otherwise be impossible, or prohibitively dangerous or expensive.”

“Conditioning human behaviour” is one of the key phrases, in my opinion. Without some really expensive VR suits which probably don’t even exist yet, you aren’t going to teach someone the physical aspects of doing the job. And that’s fine. You can teach them the muscle stuff in a safe environment. But the strength of these kinds of simulations is in teaching people what behavior is appropriate in a given situation. Until we have experienced situations over and over, we may be stuck at the point where we have to spend a relatively large amount of time assessing a situation, and then we have to think of the proper response. In a dangerous situation you don’t want to spend that kind of time. Pilots in crippled fighter planes don’t have the luxury of thinking, “I guess in this situation the best thing would be to eject. Now, where is that ejection handle?” I’ve seen video of pilots ejecting from planes in less time than it took me to realize there was a problem. Those kinds of decisions have to be “conditioned” in to you. That is where I think simulations in virtual worlds really can have some impact. Train people how to do physical things in the real world. Train them how to make the right assessments and decisions in a virtual world.

The other thing I’m disappointed in is that the article seems to indicate that the researchers are building this simulation from the ground up. Probably they have to because of their specific needs. What I really hope is that we can get to the point where researchers doing similar work don’t have to re-invent the platform but rather, can plug into an already existing platform in the metaverse, and then build their specific needs on top of that platform. Hopefully there will come a day when it’s much easier to get an island in SL, or put in their own OpenSim server online to run the simulations. As intriguing as this kind of research is, I think it would be a real shame if all simulations such as this end up running as standalone projects with no real connection to the rest of the metaverse.