Here’s an app that simulates traffic and lane merging. It’s to demo a system that will guide vehicles in a two-lane road using drum lights so they can merge at high speeds at construction spots. You get to compare a simulation “with” this system and “without” it (right now, this isn’t in production and the data’s pretty much the same, so you don’t see much difference). The idea is to show how the new merging system improves throughput. I used LightWave to model/assemble and Unity to code/render. Pretty much everything is baked/lightmapped.
I’m essentially done with it, but it’s not in production yet, so there’s still a chance to make fixes. I’d really appreciate comments/critiques, technical and non-technical (e.g., if the GUI is intuitive) for this and/or future projects.
Check it out at http://www.allurefx.com/Projects/Mergex/Mergex.html.
Looks like folks aren’t too vocal on the Unity forums when it comes to feedback. 
Anyone?
I gave it a go yesterday but wasn’t sure what you wanted feedback on? From what I could see, the simulation isn’t really interactive, you’re simply showing the difference between two vehicle control systems? I didn’t invest a lot of time playing with it, but I didn’t really see much of a difference between the two. Maybe a bit more background info, what exactly you’re trying to demonstrate/accomplish, what what we should comment on would help.
Thanks, I was looking for the usual feedback that you’d expect at the end of a project: technical/non-technical. E.g., earlier on, I had people complaining of: strange black triangles, very low fps, pops in audio, etc. - none of which I had on my system and which I ended up fixing. Do you see any issues at all? On the non-technical side, did you find the GUI intuitive or was it confusing? Stuff like that.
More importantly, do you have any expert opinions/advice on how to improve things? Since you folks are Unity experts, I was hoping for some valuable input I could use for this and/or future projects. From your comments, I guess you were looking for (or are used to looking at) games, but imagine you’re doing this project yourself. How would you have done things better?
Agreed. Looks great, but I have absolutely no idea what is going on!
I can see that there is traffic merging, but what does ‘improvement’ mean?
Really nice, though. I actually enjoyed watching it in a “I haven’t been stuck in traffic for a long time” kind of a way 
Thanks Jeff, yeah I guess this needs explanation. Actually, any ideas on how to quickly bring the viewer up-to-speed? This app is for investors/transportation people (to whom we get to explain in person), but also for the general public on the web who get to experience first-hand how the new system will feel like. We were planning to explain in text on the page as well as have a voiceover. May be also lead with some intro text on the app?
WRT the system, imagine you’re on a two-way lane and the right lane is closed ahead. Traffic will merge, but the merge process is very inefficient - do you remember how things come to a crawl at the merge only for the traffic to take off right after? This system will guide the drivers in the two lanes so they’re staggered just right to merge at high speeds. The “improvement” compares operation with this system vs. the regular merge.
Hi Sekhar,
I would try to avoid intro text as much as possible, since the chances of getting people to read text about traffic is going to be relatively low. You’re building a viewer to show the concept, so I would concentrate on making the viewer tell the story effectively instead.
Maybe a simple front end with a very brief explanation (like a single paragraph) just to ease people into the concept of a web-based simulation, then head into the main viewer for the action.
I think that once you have a few little text prompts and the voiceover in there, it will be a lot clearer. Just try to be as cinematic as possible in your telling - this is storytelling like any other, really. If you need to switch cameras to highlight a danger or really drive home a particular point, you should switch cameras.
A well thought-out smooth moving camera is also going to be more exciting that a static camera. Right now the camera seems to be just parented to the vehicle - is there any way to introduce a little movement to it? Even if the movement is a very slow zoom / dolly … just to bring it to life a little.
I hate to be sensationalist but if there are dangers or risks in either method, highlight them - use the drama of the situation to tell your story as people will always pay attention to drama! (I am NOT suggesting that you have a huge pile up with burning, exploding cars and people on fire or whatever though lol)
Those are my insane suggestions, anyway. I come from a theater and film background so I am all about the storytelling - I understand that you will be limited by budget etc., but the most important thing here is to tell the story creatively and you are already 50% doing that by creating the viewer in the first place.
Good luck!
I did actually notice this morning that in real life, the traffic in the “to be marged” lane slows down to create gaps for cars merging. In this simulation though, the cars speed up to create the gaps instead, this probably wouldn’t happen in real life, people tend to instinctively slow down when they see warning signs of “unusual” road conditions.
Thanks guys. Jeff: agreed it has to be visual (BTW, I make short films). The goal of this project though is to give a driver perspective so people feel comfortable about using the new system (it could be scary merging at high speeds), so the main camera has to stick to the dashboard view. Originally, we were planning to have the skycam (the one on the left) be roving, with dramatic views, but the space/layout wouldn’t allow it.
Still, we’ll likely have a second part to this project in which the driver will control the vehicle movement (right now, he’s just watching). I’ll be adding collision detection at that time, and it’ll be possible to drive faster/slower and bump into other cars, so there’ll be some extra action. I’ll be sure to post back here when I get to that point.
Felneymike: you’re exactly right, and that’s the point. People slow down unnecessarily/inefficiently and this system will use lights to guide you to merge at high speeds. If you hit the OFF button, you’ll see the behavior that you’re talking about (people slowing down dramatically) - except, we don’t have that data yet and are working on getting that (right now, it’s the same data for both ON and OFF). Should be ready in the next few days, will post back when that’s in.
I did actually like it. It’s awesome. How did you do that? Any tips? lol! Hope to see some more soon…
Regards,
michellefins
Disque dur multimedia
my only complaint is that MergeX appears to be slower every time…
Thanks Althearim, I created/assembled the objects in LightWave and transferred them to Unity as an FBX scene. This includes cameras and lights. All the objects I baked in LightWave. The rest of the stuff (like terrain) I created/baked in Unity. The logic is in JavaScript, and the data (for the vehicle motion) is in the JSON format. Is there anything specific you wanted to know? At some point, I guess I’ll write a blog on the more tricky aspects of this project.
Deram, the data for the WITHOUT MERGEX part is not in place yet - you’re seeing WITH MERGEX data for both the scenarios right now. I’m creating the WITHOUT MERGEX movements (in After Effects!) and will be adding that soon, like in the next few days. Will post back here then.
Neat. I was going to say I didnt see much a difference it on/off but if you dont have the data it makes sense.
So whats the idea the driver has to actually stay within the flashing lights.
I’ve updated the WITHOUT MERGEX data so you see how slow things are in real life without any system to guide you. I also added a FAST FORWARD button to speed things up if you’re impatient (it’s fun, try it).
And yeah, the MERGEX system has a series of lights on the side that will guide you to line up right for the merge. All you need to do is to keep up with the moving lights, and you’re set for a clean merge.