Need a way to get rid of directionnal bobbing

[Udonhef2bmad]

I've been working on a simple drone in order to try out what the new galaxy has to offer. (previous beta builds poofed away)

I want my drone to tilt towards the mouse since i like to pilot using left/right to strafe and leaving the directions the sensors.

The Problem:

The drone keeps bobbing left and right when adjusting its angle to the mouse. I'm looking for :

- A fast progressive stabilisation, not a never ending bobbing ride.

- Using only one directional sensor. I want my drone light.

- 4 Thrusters only. Or at least as little as you can. The kind of thruster doesnt matter as long as it's controllable.

- Logic parts on the other hand are not an issue as they'll be aboard an orbital satellite. Try to keep it simple though.

- Directional sensor must be null. That means no tolerance. I want precision (I use it backwards, yes)

- For perfection purposes, I want it to stabilise in a frictionless environment, too. That means no bobbing without air resistance, too.

 

Pretty sure most of you experienced this issue so let's settle this problem once and for all with a contest. First to fullfill my request gets a glass of water (though i won't pay the transportation costs so you'll have to go get it yourself. Also you'll have to return the glass).

That's pretty much it. Thanks for reading

[Garheardt the Black]

Those are some hefty requirements you have there. I don't think this is possible. 

Typically, drones have to trade off responsiveness with stability. In order to get high levels of both, one uses a combination of:

• Multiple sensors
• Multiple thrusters 
• Complicated logic

Your challenge asks for:

• No more than a Single sensor
• No more than four thrusters 
• Simple logic

In other words, you're asking for complicated drone behavior from an uncomplicated drone. Sadly, I doubt this is possible. 

Further, you're asking for a drone that won't "bob" but you want sensor tolerance at zero. With no tolerance, the drone will most definitely bob, the question is less about if and more about how much. 

I hate to say it, but with the current tool set, I don't think this can be done. 

 

[Udonhef2bmad]

Well yeah these are hefty requirements. I was kinda hoping some vets had already cleared this issue but apparently not.

Just clearing things up, I asked to try to keep the logic simple, implying to keep the logic as simple as possible, but since the task is that difficult, you might want to ignore that one.

I did try to do that on my own, though. I gotta admit that having both a frictionless and bob-less stablisation build is hard. Would be much easier if there were ways to interact with variables higher than 0 and 1, like the directional sensor outputting a 0-360 value you could use to gradually activate thrusters with.

That being said i do believe there is a way to do it with the current system.

 

[JMPowell86]

I think you'll find the difficulty here is not with the thruster or logic restrictions, but the sensor restriction. 

A pair of sensors have numerous capabilities that simply can't be replicated by a single sensor; without these abilities complex behaviors are very difficult to coordinate.  This is essentially why one sees so few active one-eyed species in nature.  With two directional sensors it becomes possible to detect distance and to establish closed or open-ended zones around the drone, generating a much larger possible variety of signals that can then be processed by a logic core.  Three or more sensors can open up some additional possibilities, but most complex behaviors can be achieved by two sensors.  This time, note the scarcity of three-eyed species.  The increased performance of a second sensor will almost always outweigh its mass and part cost.

You will also benefit from making use of the tolerances on your sensors.  While the 'Right' and 'Left' outputs can only be adjusted clumsily in non-90 degree increments (Devs, if you're reading this, round directional sensors, or the ability to turn the sensing portion within its block, please?), the tolerance can be easily fine-tuned, or even turned into a third output by using a pair of 'Not' gates on the sensor outputs and an 'And' gate to combine them.

Hopefully those points are of some assistance, though my ultimate conclusion is to agree with Garheardt.  Under the proposed restrictions it's likely not currently possible to build a drone that simple that doesn't move erratically.  Add a second sensor and make use of those tolerances though, and you shouldn't have too much difficulty.

[Udonhef2bmad]

Well i do know that a sensor duo works better for precision when it comes to tolerance as it narrows the tolerance zone down to basically a stripe.

But i know for a fact that a single sensor provides enough info to maneuver the drone. I mean everyone could pinpoint a direction if they are told to either turn left or right. It all comes down to how you process it.

I'm trying to avoid using a dual sensor to lighten the build and make it as compact as can be. That's some NASA level stuff right there.

Also i don't want my drone to look like one giant floaty face.

Take it that way:

• No tolerance, you only know when you're tilted left or right from the target.
• To point towards said target you'd have to spin right when the target's at your right and left when it's at your left.
• To stabilise you'd have to use less and less force on each of your correctionnal thrusts until the that force becomes null, facing the target.

I believe the main issue when it comes to this method is how to process the 3rd step and by that i mean gradually lowering the thrust.

I've been thinking since i can't access any values and don't wanna pick those wierd green boosters, to use time instead to notch down the thrust. Sill testing.

[Garheardt the Black]

It's possible to use vectored engines to orient a drone and propel it at the same time. This isn't fast, but it derives handling and thrust out of two engines. That leaves two thrusters for strafing. With this setup, you can drop sensors altogether. 

As an uncomplicated drone, there is some nuanced behavior. Expect decreased handling when you force a turn-and-burn. It also has to let off the thrusters if the mouse gets behind it. At any rate, it checks the requirements without breaking any limitations.

It wouldn't take much logic, sensing, or many additional thrusters to make this thing better at basically everything, but if you want simple AND your specified behavior, this is about as good as I can think of. 

If you're going to put weapons on something like this, I strongly recommend using zero recoil. 

  

[Udonhef2bmad]

That's actually a pretty compact way to do it i didn't think about. 

You do get minus points for that extra fuel consumption (vectors are fuel hungry) and lower thrust that get in the way.

And those are pretty major setbacks for a medium class compact drone.

That being said, this does open ways for mini recon drones.

A for effort, but i'm still not satisified.

[Garheardt the Black]

Now you're making up new rules   

You're docking merits for speed and fuel economy when:

• One large fuel tank provides 32 fuel per second
• Drone fuel consumption cannot exceed 32 fuel per second
• There was no mention of fuel economy in the challenge
• There was no mention of top speed in the challenge

Also of note, the drone I designed is most definitely a small drone. It will perform under the requirements fairly well, but if you want something more interesting, you'll need to drop some limitations. 

As for drone swarms, this was last winter:  

[Udonhef2bmad]

Well, if you're all about rules, then sorry to break it to you but one of them mentionned to use only one directional sensor.

The english language is such a beautiful thing, you can change the meaning of a sentence without touching it.

Thought you could just avoid using it? Think again.😜

 

Also these swarms look pretty neat, they're so small i can't even see  what they're made of.

[Garheardt the Black]

If we're going to nit pick, the use of "only" as an adverb can mean "No more than"

Translated: "Using only one directional sensor" = "Using no more than one directional sensor."

"No more than one" means <2

Source: https://www.dictionary.com/browse/only

---------------------

At any rate, here's a better look at the swarm drones. The parent unit follows from orbit and spawns more as desired. The individual units aren't 100% fuel efficient, but that's preferential. They're fast, hard hitting, obscenely responsive and disposable. 

This design is old by Nimbatus standards. There's plenty of room for improvement.

[Udonhef2bmad]

Alright, you got me, still i do want a directionnal sensor approach. You must really want that glass of water.

These swam drones do look awesome and simple, and it looks like you made them too, am i wrong?

[Garheardt the Black]

I did! We covered this and a lot of crazy stuff back in this thread:

If you want crazy weird inspiration, you may find some.

[Udonhef2bmad]

I love those huge ships, to some extent. I actually messed up trying to make a gigantic snake and purposely built beyond the building grids to do so.

Little did i know that it crashed the game and damaged my DirectX pilots. Had to reset my pc since i had no idea on how to repair them.

I used to build some pretty quirky stuff myself, actually managed to win 2 keys doing so. This game is like my pride as a builder.

[corona_wind]

This is a Nimbatus physics problem.  Small drones have far too little drag, while large drones have so much they're almost inertialess.  Without something like a high-drag block added, small drones will be unstable in general.  Go vote for this feature.

Jump drives to force steering to stop at the end of a turn are one way to add stability.  It's easy to overdo it.