Collision avoidance (spacecraft)

Objects launched onto suborbital trajectories will be quickly de-orbited by atmospheric drag. These include things like satellites launched on Sounding rockets designed to quickly return from orbit and rocket booster stages which expend their propellant before achieving orbital velocity. Satellites on suborbital trajectories don't usually require any intentional care on the part of the operator to ensure reentry and disposal.

The Space Shuttle external tank is designed to quickly dispose of itself after launch. The large external tank remains attached to the Space Shuttle orbiter from liftoff until when it and the orbiter are traveling at just below orbital velocity and have an altitude of approximately 113 km (70 mi), at which point it detaches and follows a ballistic trajectory quickly reentering the atmosphere. Most of the external tank disintegrates due to the heat of reentry, while the orbiter uses Reaction control thrusters to complete its orbital insertion.[5]

The vast majority of artificial satellites and space stations orbit in Low Earth orbits (LEO),[6] with mean altitudes lower than 2000 km (1200 mi). LEO satellites are close to the thicker parts of the atmosphere where safe reentry is practical because the Delta-v required to decelerate from LEO is small. Most LEO satellites use the last of their remaining onboard station-keeping fuel (used to maintain the satellite's orbit against forces like atmospheric drag that gradually perturb the orbit) to execute de-orbit burns and dispose of themselves.[7]

The ease of access for de-orbiting LEO satellites at end of life makes it a successful method for controlling the space debris risk in LEO.

Orbits with mean altitudes higher than LEO (such as Medium Earth orbits (MEO), Geosynchronous orbit/Geostationary orbit (GSO/GEO), and other species) are far from the denser parts of the atmosphere, making full de-orbit burns significantly more impractical. Few satellite designs have sufficient fuel margins to be able to afford such a maneuver at the end of their lives.

Satellites at altitudes towards the lower bound of MEO can use the "25-year rule" to decelerate with onboard propulsion so that it will fall out of orbit within 25 years, but this provision is only allowed if satellite operators can prove by statistical analysis that there is less than a 1/10,000 chance that the atmospheric reentry will cause human injury or property damage. Satellites disposed of in this fashion reenter the atmosphere in an area of the South Pacific Ocean far from inhabited areas called the spacecraft cemetery.[8]

Spacecraft orbiting at higher altitudes between LEO and High Earth orbit (HEO), most commonly in the highly specific and crowded GSO/GEO, are too far to make use of the "25-year rule". GSO and GEO require that the orbital plane be almost perfectly equatorial and the altitude be as close to a perfectly circular 35,786 km (22,236 mi), which means that space is limited and satellites cannot be allowed to stay past their useful life. Instead of decelerating for reentry, most satellites at these altitudes accelerate slightly into higher graveyard orbits where they will forever remain out of the way of interaction with operational satellites.

Historically, many multi-stage launcher designs completely expended their fuel to achieve orbit and left their spent rocket stages in orbit, as in the former Soviet Zenit family of rockets.[9] These upper stages are large artificial satellites, which depending on the orbit can take many years to reenter.

Most modern designs include sufficient fuel margins for de-orbit burns after injecting payload into orbit. SpaceX's Falcon 9 is a launch vehicle designed to minimize the effect of its upper stage on space debris. The rocket is composed of two stages, the first of which is suborbital. It reenters within minutes of launch, either intentionally using fuel reserved for stage recovery to land for reuse or is left to continue on its ballistic trajectory and disintegrate upon reentry into the atmosphere.

Falcon 9 second stages are dealt with using different techniques depending on the orbit. For Low Earth orbits, the second stage uses remaining fuel to perform a de-orbit burn and disintegrate in the atmosphere. Stages stranded in Medium Earth orbits, like Geostationary transfer orbits (GTO) and Geostationary orbit (GEO), generally don't have sufficient fuel to de-orbit themselves. GTO trajectories are designed such that the second stage's orbit will naturally decay and reenter the atmosphere after a few months, while stages from missions targeting direct insertion into GEO will remain for a lot longer.[10]

NASA uses avoidance maneuvers if the collision risk is identified sufficiently in advance and the risk is high. NASA policy for crewed spacecraft, which all have onboard propulsion, like the Space Shuttle and the International Space Station (agreed upon by all international partners) requires planning for avoidance maneuvers if the probability of collision is[4]

• >1/100,000 and the maneuver wouldn't conflict with mission objectives
• >1/10,000 and the maneuver wouldn't further endanger the crew

As of August 2020, the ISS has conducted 27 collision avoidance maneuvers since its initial launch in 1999 and is trending upwards with time. The class of debris most dangerous to the US Orbital Segment are those between 1-10 cm.[3] The population of debris in this size range is significant and difficult to track accurately with current methods, meriting further research.

These avoidance maneuvers are almost always conducted by the firing of onboard Reaction control thrusters, although some other satellite and spacecraft orientation systems like Magnetorquers, Reaction wheels, and Control moment gyroscopes may be involved. The ISS can also use the main engines of a docked cargo spacecraft – usually a Progress spacecraft or Automated Transfer Vehicle. The maneuvers slightly change the orbital trajectory and are usually conducted hours before the risk event to allow the effects of the orbital change to take effect.[4]

When two satellite operators are notified of a potential collision, one or both operators may decide to maneuver their satellite, eg. ESA & SpaceX in 2019.[11]

Recent research has developed algorithms to aid collision avoidance efforts within large satellite constellations,[12] although it is unknown whether such research has been implemented in any active constellation GNC.

Another use of a collision avoidance maneuver is to abort an automated docking, and such a procedure is built into the software that controls the docking of Automated Transfer Vehicles to the ISS. This can be initiated by the crew aboard the space station, as an emergency override, in the event of a problem during the docking.[13] This maneuver was demonstrated shortly after the launch of the first ATV, Jules Verne, and subsequently during demonstration approaches to the station which it conducted in late March 2008.

Most human-launched satellites without onboard propulsion are small CubeSats which rely on alternative devices for orientation control. At the scale of small objects like CubeSats, forces related to the large relative surface area in proportion to mass become significant. CubeSats are often launched into Low Earth orbit, where the atmosphere still provides a small amount of aerodynamic drag.

The aerodynamic drag on small satellites in Low Earth orbit can be used to change orbits slightly to avoid debris collisions by changing the surface area exposed to atmospheric drag, alternating between low-drag and high-drag configurations to control deceleration.[14]

Attempts to alleviate potential collisions are complicated by factors including if

• at least one of the offending objects lacks remote control capability due to being defunct
• at least one of the offending objects is a natural satellite, like an asteroid
• the risk event isn't predicted with sufficient time to act

All these occurrences limit strategic options for collision risk reduction in different ways. Very little can prevent the projected collision if both objects don't have control capabilities. If only one of the objects is an operational satellite, it would be the sole contributor to an avoidance maneuver, significantly cutting into or entirely using up remaining fuel reserves. The satellite may also have insufficient fuel to complete the maneuver properly, reducing its effectiveness.

Collision avoidance maneuvers require significant planning and execution time, which can be an issue if the risk isn't predicted sufficiently in advance. Spacecraft propulsion is often weak, relying on long burns to change their orbits, and the velocity change often requires a meaningful fraction of a complete orbit to produce the required effect.

For example, maneuvers commonly conducted by the International Space Station to avoid collisions often require roughly 150 second burns[15] and significant disturbances to crew operations because of the mandatory slow reconfiguration of the station's solar panels to avoid damage by propulsion devices. Roughly speaking, the estimated quickest reaction time of the ISS from normal operation is about 5 hours and 20 minutes[16] to account for the ~3 hour setup period of station reconfiguration and the ~2 hours of post-burn lead time to allow the velocity change to take effect.

Collision avoidance is a concern during spaceflight launch windows. Typically, a Collision On Launch Assessment (COLA) needs to be performed and approved before launching a satellite. A launch window is said to have a COLA blackout period during intervals when the vehicle cannot lift off to ensure its trajectory does not take it too close to another object already in space.[17]