Liftoff physics: the truth revealed!

Liftoff physics: the truth revealed!

How do the physics in Liftoff work? Well, this is the journey of an intern also determined to find out.

Liftoff’s physics?

My name is Loïc, and I’ve been an intern at LuGus Studios for roughly a month and a half now. As an engineering student, my tasks are mainly to review the way physics works in Liftoff, and to improve it, or in some cases develop whole new physical models.

But before all that, I am a Liftoff player and quadcopter enthusiast. I got into the hobby around a year ago, thanks to the very well-known 99$ build video from the Australian Youtuber UAVfutures. To be honest, it was not a very good up-to-date build, but it still got me in the air really fast and taught me some basics of drone building.

As a person obsessed with quadcopters, videogames and physics, being an intern at LuGus Studios is a crazy opportunity to work on things I’m passionate about. And as a bonus, I have to admit that the team is awesome, full of smart, creative and inspiring people! (No, I swear, I wasn’t forced into writing that).

How many physical variables does Liftoff really use ?

One of my first introductory assignment as an intern was to determine a list of factors influencing the way a quadcopter fly. Then I had to write which of these factors I thought were taken into account in-game.

So, I came up with a big list of factors from real-life flight experience, and a much smaller list for the factors I thought were used in Liftoff. I expected Liftoff to only take into account simple variables like weight, an arbitrary maximum thrust and propeller damages. Well, guess what? I was wrong.

My reasoning was that most factors I came up with would have only a little influence, and be barely noticeable anyway, so why bother using them? Actually, after some testing, it is totally obvious that enough of these factors together make a real difference in the flight sensation.

For instance, let’s take a brushless motor. It is conventionally characterized by its size, weight and Kv constant. Well, in order to compute the maximum rotation speed of a motor, Liftoff also takes into account the resistance between 2 phases, the no-load current, and the maximum power among other variables. All these variables are used only to know the maximum rotation speed of a motor with no load, which is only the very first step in the whole process of simulating a quadcopter. I let you imagine the number of parameters used to model more complex interactions!

What about the physics themselves?

Now that we know the quantity of physical variables used in Liftoff, how does the game use them to produce a realistic flight experience?

Without going in too much detail, a quadcopter flying in straight line is subject to three main forces: thrust, drag, and weight.

Weight is pretty straightforward. It is the force pushing an object toward the ground and depending on the object’s mass. Thrust is the result of each propeller spinning. It depends on the rotation speed of the motor, and the geometry of the propeller. Thrust is the reason a quad can fly. Drag, finally, is the force which opposes the quadcopter’s movement and represents the air resistance caused by viscous friction. It depends on general geometry, orientation and speed of a quadcopter.

Thrust and Drag are very complicated forces. To be able to calculate them, we would need the exact 3D files the propellers were built from and use them in a Computational Fluid Mechanics software, and we would need a wind tunnel and various measurement tools to get some specific coefficients that can’t be calculated otherwise. For a lot of reasons, this solution would be an inefficient way to do it. And even if we proceeded that way, we would have no guarantee that a quad would react in the exact same way in real life anyway – fluid mechanics can get very unpredictable really fast.

Consequently, Liftoff uses a set of empirical equations, developed thanks to large datasets acquired from manufacturers and universities. These equations give impressive results when compared to real-life measurements.

Of course, these models can always be improved, and the developers are still putting a crazy amount of work in Liftoff’s physics. That is also the reason they took me in as an intern. But the adjustments are very complex to make as every aspect of the physics depends on the others. Consequently, the changes must be iterative to make sure nothing gets broken. That is a long and tedious process, but I can assure you that some changes to the physics are on their way, so stay tuned!

Don’t hesitate to let us know if you are interested in more blog posts about Liftoff’s physics where we could go into more details about forces, equations used, or specific aspects of how a quadcopter can fly!