Details of the computer conflict on Luna-25 that led to the disaster have become known

On October 4, our country will celebrate the 66th anniversary of the beginning of the space age, that is, the launch of the first artificial Earth satellite by the Soviet Union. It so happened that just before the holiday, Roscosmos released preliminary results of work to determine the causes of the crash of Luna-25. What can you do, you can’t run away from reality. It’s good that now everything is clear to specialists and, based on this experience, they will certainly not repeat previous mistakes during the next launch of Luna-26. The MK columnist also figured out the details of the unsuccessful landing of the device.

First, the wording of Roscosmos: “It has been established that the most likely cause of the Luna-25 accident was the abnormal functioning of the on-board control complex associated with the failure to turn on the accelerometer unit in the BIUS-L device (angular velocity measurement unit) due to a possible entry into one data array commands with different priorities for their execution by the device. At the same time, the distribution of commands in data arrays is random (probabilistic) in nature.”

Let’s try to decipher what is said in the official statement, first reminding readers of how events developed in August.

The Luna-25 automatic station was launched from the Vostochny Cosmodrome on August 11, 2023 by a Soyuz-2.1b launch vehicle with a Fregat upper stage. On the flight path to the Earth’s natural satellite, it made several orbital corrections. On August 16, the station was launched into a circular lunar orbit, where it carried out scientific observations and measurements.

An abnormal “landing” occurred on August 19, 2023 due to the fact that Luna-25 slowed down too much when forming its pre-landing elliptical orbit – the engine that provided the braking continued to operate for more than the allotted time. As a result, instead of moving from a circular orbit to an elliptical one with a periselation (lunar perigee) of 18 kilometers and an aposelation of 100 km, Luna-25 switched to the so-called “open descent orbit” and simply collided with the lunar surface.

It is easier to understand the reasons for any incident by sizing up the situation, as they say, for yourself. We all live in the era of the Internet and universal computerization. Many of our problems are solved through the interaction of different processors and different computer devices with each other. For example, in order to transfer photos from a smartphone to a computer, compatible programs for their interaction must be installed on both the smartphone and the computer. But sometimes, you must admit, there are failures in this interaction.

Just such a failure occurred in a complex combination of several computers installed on Luna-25, which, of course, were required to interact. The first and most important of them is the Onboard Control Complex (BCU). This computer ensured the functioning of the entire spacecraft and its various systems: radio complex, engines, orientation system. The BKU periodically sent commands to other “computers” installed in other on-board equipment. One of them is the same BIUS-L, which, in turn, was supposed to send signals to the Onboard Control Complex about changing the speed of Luna-25, about accelerating or slowing down the flight of the spacecraft. It’s like a parking sensor on a car, which, determining the position of your car in space, begins to squeak, signaling a dangerous approach to another car.

These two Luna-25 computers described above had pre-installed programs for each of them’s own operation and programs for their interaction with each other. However, this interaction at a certain point let us down. The following happened: the on-board computer suddenly sent several commands to BIUS-L simultaneously. BIUS is like that soldier who hears from the platoon commander “right, forward, crutch on shoulder, around, running, forehead against the wall.” and doesn’t understand what to do first, what is called a system failure happened. BIUS-L carried out some commands, but ignored others.

Among the lost commands was a command from BKU BIUSu-L: “Turn on the acceleration meters of the device.” The on-board computer issued this command simultaneously with the engine turning on for braking, preparing to enter the landing orbit. And the BIUS had to report acceleration measurements to the BKU so that the BKU would give a signal to the engine to turn it off in a timely manner. As before wrote “MK”

Unfortunately, the BKU program was written in such a way that it “did not hear” the response from the BIUS – whether it understood the command of the on-board control complex. And this is where the dog was buried: the BIUS either did not receive the command, or “did not understand.” As a result, the following happened: the engines of the device worked hard to slow down, and the BIUS, without controlling the process, sent the same command to the on-board computer: “There is no slowdown from engine operation.” The most striking comparison can be made with a plane landing in fog. In this case, the pilot flies only according to the readings of the instruments, and during a normal landing one of them – the altimeter – “says”: 15 meters – 10 meters – 5 meters, etc.. And a non-working altimeter would repeat: 15 meters – 15 meters – 15 meters. And the pilot, trusting the altimeter, would have continued the flight, thinking that his altitude was still 15 m, although, in fact, it was completely different… If there had been a person on board Luna-25, he would have realized from the objective picture outside It would seem that the technology is not telling the truth, but there was no person on board Luna-25…

Consequently, a serious discrepancy arose: the BKU, upon reaching the required speed and altitude, having descended to the desired 18-kilometer orbit, had to turn off the engine. But due to the fact that there was no signal from the BIUS, he missed the moment of turning off the engine. As a result, while descending, it worked for too long, and the device literally stuck into the surface.

The most disappointing thing is that it was not the most difficult maneuver. A few days before, Luna-25 successfully completed a much more complex one, when the braking engine was also turned on, and the device carefully passed the circular lunar orbit from the interplanetary flight orbit to the Moon.

***

Many in the industry say that programmers who wrote an incorrect device control program for the BKU – that is, those same “car drivers” – should under no circumstances be fired. We must give them the opportunity to rehabilitate themselves, and then, after all, “for one beaten they give two unbeaten” – they already have, albeit sad, but no less valuable experience.

I managed to get through to the chief designer of the scientific equipment complex of the Luna-25 project from the Space Research Institute of the Russian Academy of Sciences, Igor Mitrofanov, and asked him what, in his opinion, led to the mission’s failure.

“We can’t say that everything ended in complete failure – we flew to the Moon, entered lunar orbit, confidently flew along it, photographed the Moon, observed the unique Zeeman crater on its far side,” answers Igor Georgievich. – There are no comments about the apparatus itself, about the complex of scientific equipment that our institute made. My opinion: the volume of development of the flight program was not enough. We need to help the guys overcome this misfortune and move on. As the next step in our lunar program, the Luna-26 orbiter is being prepared, followed by the next landing vehicle, Luna-27. We hope that all lessons will be learned, and next time we will confidently reach the lunar south pole.