I look forward to trying this. Improving locomotion in VR would be great.

But I feel we may be overthinking it. I believe simply tracking leg movement and having the player jog on the spot is the solution.

Having simulated this myself by holding forward on the controller while jogging on the spot, I believe it's the way forward. If you have a headset, give it a try. I admit walking doesn't feel very natural, but jogging works well.

Of course, games don't have this option with most popular current hardware (but foot trackers and whole body trackers do exist).

It would be tricky to blend/differentiate movement and jogging, but I think it's doable.

I predict that future headsets will do inside-out tracking of your entire body, including the legs (they made need a flip out camera to be far enough out from the body to do this). They've already put out a software update for the Quest that does surprisingly good controller-free hand tracking.

There was also a comment on HN a few weeks ago (I'll try to find it) where someone implemented this system and said it worked well.

> Of course, games don't have this option with most popular current hardware

There's a great low tech solution - pair your mobile phone and leave it in your pocket. The accelerometer data is more than good enough to detect jogging.

There was an app for Steam VR that did this and I recall it worked quite well.

EDIT: here: http://pocketstrafe.com/

treadmill solutions seems to be the way the industry is going. The Kat Walk C is a kickstarter project that actually works pretty well, and relatively small:

https://www.youtube.com/watch?v=sqchs9xMLCw

I'm really curious to try this. It's interesting how we tend to veer off course when trying to walk straight, and cool to see an idea of how to exploit it.

I used to do gymnastics and as part of that I trained doing long series of cartwheels along a marked line. I managed to get good at doing it without deviating from that line. I looked straight ahead and it went fast enough that everything was a blur. However, I still couldn't do it with my eyes closed.

Hi, I'm the lead on walking dream. There have been a couple of other games I've seen that have implemented version of this, though using very different approaches. However, most were closer to tech demos, without a full game experience.

I will be very curious what the minimum area requirement for the game is. I can't imagine redirected walking being very convincing in the Quest's minimum 6.5 x 6.5 ft square.

I've played around in Tea For God (https://sidequestvr.com/app/65/tea-for-god-under-development) which does this on the quest, and it works really well, and I didn't even use the full 6.5'x6.5' (in my case ~6'x4') possible. Within a minute I lost all sense of where I was in the real world.

Tea For God does not use redirected walking. It procedurally adopts game space to available room space, and uses clever mechanisms like non-euclidian geometry and various elevators to allow for huge worlds. Still, movement is 1:1 mapped between room and game.

Redirected walking means distorting the mapping when moving or rotating, to fit any game world into limited playing space. In research it's always applied in larger spaces, x5 times bigger than living rooms. If you distort moving or turning too much, users lose balance or become nauseous.

Here's a decent intro paper. 15 Years of Research on Redirected Walking in Immersive Virtual Environments https://illusioneering.cs.umn.edu/papers/nilsson-cga2018.pdf

Hi, I'm the lead dev on walking dream. It's hard to say right now what the "minimum area requirement" is, because there's an inverse correlation between area size and nausea (since more "redirecting" is needed in smaller spaces) However, I think it works pretty well in 9x9ft so far (but we need to see how it works for other testers)

I agree that 6.5ft^2 is probably too small for a good RDW experience based on the gain numbers I'm familiar with. The tracked volumes of the experiments I was a part of were all on the scale of ~15-20 ft square. (I'm not a researcher, really, but I was a developer at a lab that did some RDW research -- for instance, https://dl.acm.org/doi/abs/10.1145/2931002.2931018). There are some tricks that the literature suggests could help (forcing the user to turn >360 degrees in place, exploiting change blindness to get people to accept non-euclidean spaces, etc) but I believe it's not controversial to assert that the user is going to realize they're not walking straight if someone tries to get them to walk a circle with radius ~3ft with traditional RDW cues!

I'm the lead dev on Walking Dream: No doubt the player will "notice" they are being redirected in a small space, the question is if it's bad enough to induce nausea. Our goal is to be able to tell the player "you can just walk wherever you want in the game, but you'll notice some weird rotational shifting, however you won't get nausea". It's a lower bar that can still lead to fun gameplay.

(2m × 2m if you're curious)