← Back

Why is space software special?

What is Point of No Return in space software? Why you need to do a bit extra for space software compared to ordinary software?

Slavo Petrik   -   July 3rd, 2023

space software

Once launched, your software will have to live all its faults.

Want to work on space software? You better know this: the space software has a point of no return. Wandering what that is?

My first space software project was for the Slovak Ministry of Education. A small CubeSat called SkCube 1. A 10x10x10 cm satellite built in 2010. Anyhow, it operated quite a lot of software. Sensors control, communication control, attitude control algorithms – the list was long.

Indeed I knew that there is a difference between the software in your washing machine and the software in the satellites. Why otherwise would the space industry accept only the best developers out there?

What I didn’t know was what extra things to do for the space software to make sure it would run. Is it more testing? Is it a better design? Is it different architecture?

Since then, I contributed to many other space projects. One thing was the same in all of them. The satellite’s lifetime was always split by the particular point I call The Point of No Return.

The Point of No Return

The point of no return seats right there – in the scary spot between the last verification campaign in the lab and the integration into the launch container in the clean room. After that, you will hopefully hear your system from the orbit.

Why do I call it the Point of No Return? Because once you hand over your space system to the launch operator, that’s it. Your space system will have to live with all its faults and poorly designed parts, and that’s the biggest fear of the satellite developers…

You can’t update your software in orbit (at least not without a significant risk to your $300K+ mission). And even if you manage to update it, you will experience 90 minutes of terror waiting for the satellite to finish the orbit, establish a communication session, and tell you that the restart after the software update went well.

Let’s look at what is to the left and right of The Point of no Return.

On the left side of the Point of no Return, you do “everything you can” to ensure the satellite will provide you with data collection and other functions. That “everything you can” in the previous sentence is what makes the space software special. After the Point of No Return, there are no fixes, updates, or patches. There is only a successful mission or a long string of creative sessions on how to run your satellite in degraded mode with limited functions and limited publicity.

In the following articles, I will write about the left and right sides of the Point of No Return. For now, please internalize the concept of the Point of No Return – it will motivate you to come with extra creativity when designing and testing the space systems.

You only get one chance to run your satellite in the space. Your mission will either be successful or not – depending on how much extra you did before the Point of No Return compared to the normal software development you learned in school. And I can tell you that more testing will generally not improve your software as much as you hope… the trick is elsewhere.