Could Something Knock an Asteroid Off Course and Send It Straight Toward Earth?

Key Takeaways

• Passing interstellar bodies exert far too little gravity to redirect typical asteroids.
• Direct hypervelocity impacts can alter orbits, but such interstellar collisions are exceedingly rare.
• Asteroid collisions can create fragments that later migrate into Earth-crossing orbital paths.

DART Provides a Measured Orbit-Change Benchmark

On September 26, 2022, a spacecraft weighing about 579 kilograms struck the asteroid moonlet Dimorphos at 6.145 km/s and measurably changed its motion. NASA’s Double Asteroid Redirection Test (DART) provided an unusually useful experiment for assessing asteroid orbit disruption because the spacecraft’s mass, speed, impact location, and resulting orbital change were measured rather than inferred from an ancient natural collision. The DART technical record gives an impact energy of about 10.94 gigajoules and an orbital-period reduction of roughly 33 minutes.

Dimorphos is about 150 to 170 meters across and orbits the larger asteroid Didymos. DART did not send Dimorphos onto a radically different orbit around the Sun. It changed the moonlet’s orbit around Didymos. More recent NASA analysis found that material ejected during the collision increased the momentum transferred beyond the momentum carried by the spacecraft itself. The same impact also produced a tiny measurable change in the Didymos system’s orbit around the Sun, about 0.15 seconds in orbital period according to NASA’s later DART analysis.

That distinction establishes an important scale for the scenario considered here. Asteroids certainly can have their orbits altered by impacts. Humanity has demonstrated the mechanism directly, and natural asteroid collisions have operated throughout Solar System history. The question is how large a velocity change is required to transform a particular asteroid from its existing orbit into one that intersects Earth’s orbital region.

The answer depends overwhelmingly on where the asteroid begins.

An object already approaching Earth’s orbit may need only a small velocity change to alter the location or timing of a later close approach. A main-belt asteroid on a relatively circular orbit between Mars and Jupiter faces a different problem. Throwing such an object directly onto an Earth-crossing orbit requires an orbital-energy change that can be thousands of times larger than the change needed to shift the arrival time of an asteroid already passing close to Earth.

NASA’s planetary-defense DART program illustrates the opposite strategy. A defender does not ordinarily need to move a dangerous asteroid by thousands of kilometers at the moment of intervention. With enough warning time, a tiny velocity change accumulates into a large positional difference years later.

Natural impacts work under the same physics but without favorable timing, direction, target selection, or warning. An impact from another asteroid or from an interstellar object can increase orbital energy, decrease it, change inclination, alter rotation, fragment the target, or produce some combination of these effects. Only a small subset of collision geometries would move material toward an Earth-crossing configuration.

New Space Economy’s planetary-defense technology overview discusses intentional deflection technologies in this same physical framework. Natural asteroid collisions amount to uncontrolled kinetic impacts, with the added complication that energetic events may destroy rather than neatly redirect their targets.

Asteroid Orbit Disruption Starts With Orbital Geometry

An Earth-crossing asteroid is not simply an asteroid that has moved closer to the Sun. Its orbit must extend into the radial region occupied by Earth’s orbit, and its three-dimensional geometry determines whether the two orbital paths can actually intersect.

NASA distinguishes Earth-crossing asteroids from the broader near-Earth asteroid population. A near-Earth asteroid generally has a perihelion, its minimum solar distance, below 1.3 astronomical units (AU). An astronomical unit is the average Earth-Sun distance, about 149.6 million kilometers. An asteroid can satisfy the near-Earth definition without physically crossing Earth’s orbit.

Consider an intentionally simplified case. Suppose an asteroid follows a nearly circular orbit at 2.5 AU, close to the location of Jupiter’s 3:1 orbital resonance in the main belt. Its orbital speed is about 18.84 km/s.

For an instantaneous velocity change applied in exactly the favorable direction at 2.5 AU to place the asteroid onto an elliptical orbit with an aphelion of 2.5 AU and a perihelion of 1 AU, the required reduction in tangential velocity is roughly 4.6 km/s.

That is an enormous velocity change for an asteroid.

Reducing the perihelion only to the conventional 1.3 AU near-Earth boundary still requires roughly 3.3 km/s under the same simplified assumptions. These calculations neglect inclination changes, planetary perturbations, rotational effects, fragmentation, and resonance evolution, but they establish the scale of a direct orbital transfer.

A collision does not usually provide an asteroid with such a clean impulse. The impact energy produces excavation, heating, deformation, seismic motion, ejecta, rotation changes, and possibly fragmentation. Only part of the projectile momentum becomes useful center-of-mass motion of an intact target.

The picture changes dramatically for an asteroid already close to a dynamically unstable region. The main belt contains resonances created by repeated gravitational interactions with Jupiter and other planets. NASA’s main-belt distribution data show prominent Kirkwood gaps associated with resonances such as 3:1, 5:2, and 2:1 with Jupiter.

A collision does not have to push a main-belt asteroid directly from 2.5 AU to Earth’s orbit. It may need only to move a fragment into a nearby resonance. Planetary gravity can then increase orbital eccentricity over much longer periods, lowering perihelion until the object becomes Mars-crossing and eventually near-Earth.

This indirect path changes the probability calculation. A kilometer-scale parent asteroid far from a resonance is difficult to turn directly into an Earth-crosser. Fragments produced close to a strong escape route can require much smaller initial orbital changes.

Another distinction concerns crossing versus collision. An asteroid may acquire an orbit that crosses Earth’s orbital distance yet miss Earth on every passage because Earth arrives at the intersection at a different time. Inclination can place the asteroid above or below Earth’s orbital plane at the relevant solar distance. The orientation of the orbit also changes under planetary perturbations.

Becoming Earth-crossing is consequently an intermediate condition, not an impact prediction.

Passing Interstellar Objects Produce Almost No Gravitational Deflection

Interstellar objects are now an observed population rather than a hypothetical class. Astronomers identified 1I/’Oumuamua in 2017, 2I/Borisov in 2019, and 3I/ATLAS in 2025. NASA describes 3I/ATLAS as the third confirmed interstellar object observed passing through the Solar System.

Their speeds matter greatly for the proposed asteroid-disruption scenario.

NASA gives ’Oumuamua an incoming speed relative to the Sun near 26.4 km/s before solar gravity accelerated it during passage through the inner Solar System. Research on 3I/ATLAS found an interstellar excess speed close to 58 to 60 km/s. Such bodies pass local asteroids very quickly unless their directions happen to align.

Gravity from a small interstellar body acts for only a short time during a high-speed encounter. An approximate velocity impulse from a fast gravitational flyby can be written as:

Δv ≈ 2Gm/(bv)

Here, m is the mass of the passing object, b is the closest-approach distance, v is relative speed, and G is the gravitational constant.

Take a hypothetical spherical interstellar object 100 meters in diameter with density near 2,000 kilograms per cubic meter. Its mass would be about 1 billion kilograms. Suppose it passes only 1 kilometer from an asteroid at 30 km/s without colliding.

The resulting gravitational velocity change would be only a few billionths of a meter per second.

Even changing the assumed mass or encounter distance substantially does not make an ordinary‘Oumuamua-scale object a meaningful gravitational perturber of a kilometer-scale asteroid. The projectile’s gravity is simply too weak and the encounter too fast.

This is different from a planetary flyby. Earth, Jupiter, or another planet has enough mass to reshape an asteroid orbit substantially during a close encounter. A small interstellar comet does not.

For comparison, 3I/ATLAS was traveling through the planetary region at high speed, yet NASA determined that it posed no danger to Earth. Its discovery does establish that extrasolar debris enters the Solar System. NASA’s treatment of interstellar-object discoveries lists three confirmed macroscopic examples as of August 2026.

Estimates of the unseen interstellar population remain uncertain because the observed sample is tiny. A widely cited 2018 analysis inferred a number density near 0.2 ’Oumuamua-like objects per cubic AU. NASA has also described objects of roughly this observational class as potentially passing through the inner Solar System on the order of once per year. Those estimates concern passage through a vast astronomical volume, not close encounters with individual asteroids.

The difference in scale is enormous. One AU is almost 150 million kilometers. An asteroid that is 1 kilometer across occupies a microscopic collision target compared with an AU-scale region.

A stellar-mass interstellar visitor or an unbound planet would be different because its gravity could perturb many Solar System objects from a distance. No such object is known to be approaching the asteroid belt, and that scenario belongs to a completely different probability class from the confirmed interstellar comets and asteroid-like bodies represented by ’Oumuamua, Borisov, and 3I/ATLAS.

For known small-body interstellar objects, gravitational flyby disruption of a normal asteroid can be treated as physically real in principle but negligible in practice.

Direct Interstellar Impacts Can Change Orbits but Are Extremely Rare

Impact changes the calculation because a collision transfers momentum directly.

Imagine again a 1-kilometer asteroid and a 100-meter interstellar projectile with roughly the same density. Their diameter ratio is 10 to one, giving a mass ratio near 1,000 to one. If the smaller object strikes at 30 km/s and momentum transfer is treated in a simple perfectly inelastic approximation, the target’s center-of-mass velocity change would be on the order of 30 m/s before accounting for fragmentation and impact ejecta.

That is millions to billions of times larger than the gravitational kick produced by a close non-impacting flyby.

It is still far below the roughly 4.6 km/s change calculated for directly transferring a circular 2.5 AU asteroid onto an orbit with a 1 AU perihelion. A projectile large enough to produce that change through momentum alone would represent a substantial fraction of the target’s own mass. At interstellar speeds, such an event would resemble a destructive collision rather than a tidy redirection.

Fragmentation then becomes central. Some pieces could receive higher velocities than the parent body’s center of mass. Their directions would differ. A few might move toward dynamically unstable regions, others away from them, and much of the debris could remain on heliocentric orbits close to the parent’s former orbit.

The probability of the collision itself is extraordinarily small.

A useful order-of-magnitude calculation can be made using the 2018 estimate of approximately 0.2 ’Oumuamua-like objects per cubic AU. The estimate is uncertain and should not be interpreted as a precise measurement of the current interstellar population.

Assume an interstellar speed of 26 km/s, a 1-kilometer target asteroid, a roughly 100-meter projectile, and only the geometric collision cross section. Under those assumptions, the collision rate for one target is roughly 5 × 10^-17 per year.

That corresponds to a characteristic waiting time of around 20 quadrillion years for that individual asteroid.

The figure should not be read literally because the interstellar size distribution, speed distribution, gravitational focusing by the Sun, target orbit, projectile composition, and observational population estimate all carry uncertainty. Its value is in the scale comparison. Even changes of several orders of magnitude would leave a 100-meter-class interstellar projectile striking a particular kilometer-scale asteroid extraordinarily uncommon compared with the 4.5-billion-year age of the Solar System.

Larger target asteroids present greater cross sections, but they are less responsive to a given projectile because their masses increase rapidly with diameter. Smaller targets are easier to move or fragment, yet their collision cross sections shrink.

Smaller interstellar projectiles are presumably more numerous, although their population has not been measured well enough to extend the 100-meter-class estimate confidently across many size decades. Their momentum also falls rapidly with diameter. Reducing projectile diameter by a factor of 10 cuts its mass by about 1,000 for similar density.

A 10-meter projectile hitting a 1-kilometer target at 30 km/s would produce a simple momentum-based center-of-mass velocity change of only around 0.03 m/s. That may matter for an asteroid already near a resonance or a sensitive future close encounter, but it does not directly transport a normal main-belt asteroid to Earth’s orbit.

The main effect of interstellar impacts on asteroid hazard is consequently likely to be statistical and extremely small rather than a mechanism that routinely creates new Earth-crossers.

Asteroid-on-Asteroid Collisions Are a More Plausible Source of Orbital Change

The Solar System already contains a dense population of potential asteroid projectiles in the main belt. Natural collisions between those objects have shaped the belt for billions of years.

NASA estimates that the main asteroid belt contains roughly 1.1 million to 1.9 million bodies larger than 1 kilometer, plus much greater numbers of smaller objects. Evidence of collisions appears in asteroid families, craters, fragments, disrupted bodies, binary systems, and rubble-pile asteroids.

Research on the collisional evolution of the belt has estimated that a roughly 1-kilometer main-belt asteroid has a characteristic catastrophic-disruption lifetime near 440 million years. A roughly 500-meter object can have a shorter lifetime near 200 million years, depending on the adopted impact population and disruption model. Modern numerical work on Bennu and Ryugu cites these approximate timescales when examining their likely collisional ancestry.

Those times are long on human scales but entirely relevant across Solar System history. Hundreds of millions of years allow collisions to repeatedly produce new asteroid fragments.

The mechanism also differs from the imagined case of a large asteroid simply being knocked onto an Earth-crossing orbit as a single intact object. Catastrophic collisions usually distribute the parent’s mass among fragments with a spread of velocities. Those fragments begin on heliocentric orbits similar to the parent body’s orbit, but small changes in semimajor axis, eccentricity, and inclination can determine whether a fragment lies close to a resonance.

This process is one reason asteroid families matter. Members share a common collisional origin but spread dynamically over time.

NASA’s historical work on near-Earth-object science describes how resonances involving Jupiter and Saturn can increase asteroid eccentricities until objects leave the main belt and enter the inner Solar System. The history of NEO research describes the 3:1 resonance and the ν6 secular resonance as important escape paths.

Collisions contribute in two ways. They create fragments, and they give those fragments slightly different initial orbits. Thermal forces from sunlight then alter the semimajor axes of smaller bodies over long periods, allowing some fragments to drift toward resonances.

This route does not require one spectacular collision that immediately creates an Earth-threatening object. A far smaller orbital displacement can place a fragment into a region where repeated planetary perturbations perform most of the later orbital evolution.

That makes asteroid-on-asteroid collisions much more relevant to the production of the near-Earth population than interstellar collisions.

The relationship is still probabilistic. Most collision fragments do not become near-Earth asteroids. Many remain in stable main-belt orbits, collide again, drift in other directions, or are removed through other dynamical processes.

A collision close to an escape resonance has a different probability from one deep inside a stable portion of the belt. Fragment size matters because thermal drift is size dependent. Spin orientation matters because it influences the direction of that drift. The initial ejection velocity matters, as does the orientation of the impact relative to the parent’s motion.

New Space Economy’s asteroid impact risk discussion places such orbital evolution within the wider planetary-defense problem. Hazard calculations concern where an object will actually be in the future, not simply whether some event has changed its orbit.

Resonances Can Amplify Small Orbital Changes Over Long Periods

A small collision does not need to supply all the energy needed for an Earth-crossing orbit if it places an asteroid or fragment into a dynamically unstable region.

The 3:1 mean-motion resonance with Jupiter near 2.5 AU provides a clear example. An asteroid in this region completes about three solar orbits during each Jupiter orbit. Repeated gravitational interactions occur with a regular pattern. Over time, the resonance can increase orbital eccentricity enough to move objects out of stable main-belt configurations.

The ν6 secular resonance near the inner belt operates differently but can also raise eccentricity and deliver material toward planet-crossing orbits.

Research tracing the origins of near-Earth asteroids identifies several escape regions, including ν6, the 3:1 resonance, the 5:2 resonance, the 2:1 resonance, and source regions associated with the Hungaria and Phocaea populations. The exact proportions depend on asteroid size and the population model.

Thermal forces help feed those escape routes. The NASA Yarkovsky explanation describes how sunlight absorbed and reradiated by a rotating asteroid produces a tiny recoil force. Over millions of years, that force can measurably alter the orbital distance of smaller asteroids.

The measured Yarkovsky effect on asteroid Golevka provides a useful sense of scale. NASA reported that a tiny thermal force shifted the predicted position of the asteroid by about 15 kilometers over 12 years. Over much longer periods, thermal drift can move main-belt fragments into resonances that change their eccentricities far more strongly.

This creates a sequence:

A parent asteroid is disrupted or cratered. A fragment receives a modest velocity change. Its new orbit places it closer to a resonance or thermal drift gradually carries it there. Planetary perturbations increase eccentricity. The fragment becomes Mars-crossing and may later enter the near-Earth population.

No single stage needs to provide the entire 3 to 5 km/s orbital-energy change that a direct transfer from a circular main-belt orbit would require.

This is the strongest physical route connecting asteroid collisions with future Earth-crossing objects.

An interstellar collision could theoretically begin the same sequence. The origin of the projectile would not matter once momentum had been transferred. A collision involving an indigenous main-belt asteroid, an interstellar asteroid, or an interstellar comet could create fragments that enter suitable orbital regions.

Probability distinguishes the cases. Main-belt projectiles occupy the same region for billions of years and number in the millions at kilometer scales, with vastly greater populations at smaller sizes. An interstellar projectile passes through once at high speed.

The ordinary collisional population consequently dominates the long-term production mechanism.

Small changes also matter for asteroids already in near-Earth space. Planetary close approaches can act as gravitational scattering events, sometimes changing an asteroid’s orbit far more than a small collision would. This makes long-term prediction increasingly sensitive when an asteroid experiences a close planetary encounter.

NASA’s Sentry system addresses this by propagating possible future orbits and searching for Earth-impact solutions. The Sentry monitoring system examines uncertainty in orbital solutions rather than assuming one perfectly known path.

That approach would remain applicable if an asteroid experienced a naturally induced orbital change. Fresh observations would reveal that its measured position no longer matched the earlier orbital solution, and the orbit could be recalculated.

Earth-Crossing Does Not Mean Earth Impact

A newly Earth-crossing orbit sounds more dangerous than it often is.

Earth occupies a tiny fraction of its 940-million-kilometer orbital path at any instant. For an asteroid to strike the planet, the asteroid and Earth must reach the orbital intersection at almost exactly the same time. The asteroid’s vertical position relative to Earth’s orbital plane must also be compatible with an encounter.

An orbit can consequently cross Earth’s orbital distance for thousands or millions of revolutions without producing an impact.

Planetary-defense calculations distinguish orbital classification from collision probability for this reason. A near-Earth object may have an orbit that brings it close to Earth yet possess no meaningful impact probability over the period for which its orbit can be predicted reliably.

Asteroid 2024 YR4 provided a recent example of another aspect of the problem. Its physical orbit did not suddenly change when its calculated Earth-impact probability rose and later collapsed in early 2025. Additional observations narrowed the uncertainty in its predicted 2032 position. NASA’s analysis shows the calculated probability falling to about 0.004% once enough observations excluded Earth from the relevant uncertainty region.

That distinction matters for a hypothetical collision-induced orbit change. If astronomers observe the collision and obtain measurements afterward, the new orbit can be determined directly. If the event occurs unseen, discovery may come later when surveys identify the object or its fragments in their altered orbits.

Current planetary-defense systems are designed around the final observable hazard rather than the origin of the orbital change. NASA’s Center for Near Earth Object Studies and ESA’s Near-Earth Object Coordination Centre calculate future positions from astrometric observations and update risk assessments as new measurements arrive.

The source of the velocity change might matter scientifically, but the impact prediction still comes from the resulting orbit.

A collision that created hundreds of fragments introduces another complication. Large fragments could have somewhat different orbits and therefore require separate tracking. Small debris would be more difficult to detect, although smaller objects also produce less damage if they encounter Earth.

The physically relevant probability chain is consequently narrower at every stage. An interstellar object must encounter the Solar System. It must pass through the part of the asteroid population containing a suitable target. It must collide rather than miss. The projectile must have enough mass and an appropriate impact geometry to produce the required change or useful fragments. Those fragments must enter a route that reaches the near-Earth region. Their orbital geometry must become Earth-crossing. Earth and the object must then arrive at an intersection simultaneously.

Multiplying rare conditional events produces a probability far below the probability associated with the already recognized natural supply of near-Earth asteroids from the main belt.

The Most Plausible Version of the Scenario

The scenario becomes much more physically credible when it is modified from “an interstellar object knocks a main-belt asteroid directly toward Earth” to “a collision changes the orbit of a fragment that was already close to an unstable dynamical route.”

Consider a parent asteroid near the 3:1 Jupiter resonance. A collision produces fragments traveling tens or hundreds of meters per second relative to the parent, depending on impact energy and fragmentation physics. Some fragments move inward in semimajor axis and others outward.

A subset may enter the resonance quickly. Jupiter’s repeated gravitational perturbations can then increase their eccentricities, producing planet-crossing paths without the original collision delivering kilometers per second of direct velocity change.

The projectile in that scenario is much more likely to be another Solar System asteroid than an interstellar object.

Another plausible case involves an asteroid that is already a near-Earth object. A collision producing a velocity change of centimeters or meters per second might materially alter a close approach decades later. The required impulse can be small because the issue is arrival timing rather than moving the asteroid from the main belt into Earth’s region.

DART demonstrated exactly why timing matters. A modest velocity change applied years before a predicted encounter can produce a large displacement when the encounter date arrives. New Space Economy’s deflection strategies overview describes intentional use of this principle for planetary defense.

Natural impacts can accidentally operate in either direction. A collision might reduce future Earth risk by shifting an asteroid away from a close approach, or it might create a future encounter that was absent from the previous orbital solution.

No physical law favors Earth-directed outcomes.

That symmetry is frequently lost in hypothetical treatments of asteroid disruption. Space is enormous, Earth presents a small target, and most randomly directed impulses move an asteroid onto an orbit that still misses the planet.

Fragmentation makes an Earth-crossing outcome easier for some debris because the collision creates many independent orbital paths. It simultaneously spreads the original mass among smaller bodies, changing the consequences of any later impacts.

A very energetic collision can also remove the concept of a single “new orbit.” The parent body may cease to exist as one gravitationally coherent asteroid. Each large fragment acquires its own orbit.

This is why ESA’s Hera mission has scientific relevance beyond DART itself. As of August 26, 2026, Hera is approaching Didymos for a planned rendezvous later in 2026. Its measurements are intended to constrain Dimorphos’s mass, internal structure, surface properties, and the physical outcome of DART’s collision. Those quantities govern how kinetic energy and momentum translate into asteroid motion and ejecta.

Natural collisions involve the same physics at different scales and geometries.

Summary

Asteroids can have their orbits changed by impacts, and some collision fragments eventually reach Earth-crossing orbits. That process already forms part of the accepted explanation for how material escapes from the main asteroid belt and enters near-Earth space.

A passing interstellar asteroid or comet of the type represented by ’Oumuamua, Borisov, or 3I/ATLAS cannot realistically redirect a normal asteroid through gravity alone. Its mass is too small and its passage too fast. A direct impact can transfer vastly more momentum, but the geometric probability of an interstellar object striking a particular asteroid is extraordinarily low.

Using an ’Oumuamua-like number-density estimate and a simplified 1-kilometer target calculation gives a per-target collision frequency near 10^-17 events per year for a roughly 100-meter-class interstellar projectile. The assumptions are uncertain, but the result is so small that plausible revisions do not turn the event into a conventional planetary-defense hazard.

A direct collision also faces an energy problem. Moving a circular 2.5 AU asteroid immediately onto an orbit reaching 1 AU requires roughly a 4.6 km/s velocity change under favorable simplified geometry. A 100-meter projectile striking a 1-kilometer target at 30 km/s produces only about 30 m/s of center-of-mass velocity change in a basic momentum estimate.

The more realistic pathway uses dynamical amplification. An impact creates fragments or modest orbital changes near a resonance. Thermal forces and planetary perturbations then move some fragments toward Mars-crossing and near-Earth configurations over long periods.

Ordinary asteroid-on-asteroid collisions are vastly more plausible initiators of that sequence than collisions with interstellar visitors. Main-belt collision models give disruption lifetimes of hundreds of millions of years for kilometer-scale bodies, providing ample time for repeated fragmentation throughout Solar System history.

None of this means every newly Earth-crossing fragment represents an Earth-impact threat. Crossing Earth’s orbital distance is one geometric condition among several. Actual impact requires the orbital paths, nodes, timing, and future planetary perturbations to align.

For planetary defense, the most important implication is observational. The Solar System continually changes through collisions, thermal forces, resonances, and planetary encounters. An asteroid catalog is consequently not a frozen inventory. Persistent sky surveys, orbit refinement, and impact monitoring remain necessary because natural processes can create new near-Earth objects even when the original parent bodies occupied the main belt.

The interstellar-object scenario is physically possible, but it does not appear to represent a meaningful additional source of Earth-impact risk compared with the ordinary dynamical and collisional processes that already supply the near-Earth asteroid population.