Steering a warp drive without exotic matter

Abstract: A useful warp drive must change its velocity, yet known positive-energy constructions are static or constant-velocity, while accelerating ones require exotic matter; Bobrick and Martire noted that warp drives ``do not have any natural way of changing their velocities.'' First, a conservation law settles the principle: for any asymptotically flat drive with a confined dominant-energy source and standard peeling, the Bondi--Sachs four-momentum changes only through radiation to null infinity, so the drive cannot steer without radiating. Second, we construct a positive-energy spacetime that saturates this bound. Prescribing the passenger worldtube, we take the exact Kinnersley photon rocket as exterior and solve for the matching timelike shell; the exterior and steering law are exact, while the shell is certified perturbatively and numerically. The mechanism is photon-rocket recoil on a tidally protected, exactly flat cavity; the warp drive is the decoupled flat interior, not a warp field. The exterior energy conditions reduce to $n<sup>2\ge0$, and steering obeys $-\dot m\ge 3m|a|$, paid for by Bondi mass loss. The shell satisfies the surface dominant energy condition for $2m/R<24/25$; admissible accelerating shells exist near the Schwarzschild--Minkowski anchor and, for slow burns, as time-evolved spacetimes. A Buchdahl-type frontier $a_{\max}R\le g(2m/R)$ caps the acceleration between a numerical lower bound $\sim!0.2$ and a closed-form ceiling ${1\over2}(1-2m/R)$. On the gravitational-wave--silent class, the optimal maneuver is the Damour dipole. The wall is marginally stable, with strict stability available at no energy-condition cost; fully dynamical flux-coupled stability remains open. The drive is causal, subluminal, and energetically costly but positive-energy: steering is a problem of energy budget, not exotic matter.

• The paper constructs a subluminal accelerating warp drive using only positive-energy matter, matching a flat passenger cavity to a Kinnersley–Robinson–Trautman photon-rocket exterior and an admissible shell.
• The paper proves that any confined positive-energy drive must radiate to change its Bondi velocity, and derives the steering law −ṁ ≥ 3m|α|, requiring about 80% mass loss to reach 0.5c and over 98% to reach 0.9c.
• The paper establishes perturbative and numerical shell-existence results, bounds acceleration through compactness, and identifies unresolved challenges including nonlinear back-reaction, flux-coupled stability, and rigorous frontier certification.

This paper constructs, and certifies, an accelerating warp drive built entirely from positive-energy matter. The construction inverts the usual design logic: rather than prescribing a metric and accepting whatever stress-energy the Einstein equations return (the route that forces exotic matter in the Alcubierre program), the author prescribes a passenger worldtube and its flat cavity, adopts the exact Kinnersley–Robinson–Trautman photon rocket as the exterior, and solves for the matching timelike shell. The organizing result is a classical conservation law read as a propulsion no-go, of which the explicit spacetime is a saturating instance.

The no-reactionless-steering theorem

The capstone is a model-independent theorem: for any asymptotically flat spacetime with a spatially confined material source satisfying the dominant energy condition and admitting Bondi–Sachs peeling, the Bondi four-momentum evolves only through the flux radiated to null infinity,

$$\frac{dP_B^\mu}{du} = -\frac{1}{4\pi}\oint\Big[|\mathcal{N}(u,\Omega)|^2 + 4\pi\, n^2(u,\Omega)\Big]\hat\ell^\mu\, d\Omega,$$

where $\mathcal{N}$ is the Bondi news and $n^2$ the outgoing matter flux. Since both integrands are non-negative and contracted with a future-directed null vector, $P_B^\mu$ is constant on any flux-free segment: a confined positive-energy drive cannot change its velocity without radiating. The proof composes classical ingredients — the Dray–Streubel/Geroch–Winicour charge, the Bondi–Sachs flux law, and the Nester–Witten positivity of Bondi mass, which closes the "borrow momentum by driving the mass negative" loophole. The paper is careful about what is and is not new: the flux balance itself is classical; the contribution is its reading as a no-go, its separation from the center-of-mass "swimming" effect (which displaces position at fixed $P_B^\mu$), and the explicit positive-energy saturating instance. The instantaneous identification of the center of mass with the $\ell=1$ mass-aspect moment is conceded to be supertranslation-ambiguous in general; it is discharged exactly here because the Kinnersley exterior supplies the closed-form worldline $P_B^\mu = m(u)\,v^\mu(u)$.

The law is verified numerically to machine precision. The Bondi balance closes two independent ways — kinematically from $P_B^\mu = m v^\mu$ and from the curvature-derived flux — to $\sim 10^{-15}$, and a model-independent extractor confirms it on three spacetimes (Kinnersley, Schwarzschild, Vaidya). The Newman–Penrose peeling hypotheses are certified directly: $\Psi_2 \sim r^{-3}$ with fitted slopes $\mathcal{N}$0, $\mathcal{N}$1, $\mathcal{N}$2, while a deliberately non-silent linearized-wave control reproduces $\mathcal{N}$3, confirming that the Damour dipole's vanishing $\mathcal{N}$4 is genuine gravitational-wave silence rather than an insensitive instrument.

The construction and the control law

The exterior stress-energy is pure outgoing null dust, $\mathcal{N}$5, algebraically Hawking–Ellis Type II. Because $\mathcal{N}$6 is null, every contraction with a causal vector is proportional to $\mathcal{N}$7, so all bulk energy conditions (NEC, WEC, SEC, DEC) collapse to the single inequality $\mathcal{N}$8. The field equations fix $\mathcal{N}$9; writing the steering dipole as $n^2$0, with $n^2$1 equal to the covariant proper acceleration, positivity for all angles yields the closed-form control law

$n^2$2

The saturated integral is the Tsiolkovsky budget in Bondi form, $n^2$3. The factor 3, against the ideal collimated photon rocket's $n^2$4, is the price of the mandatory broad Kinnersley dipole exhaust, which delivers only the $n^2$5 projection of its radiated energy as net momentum. The paper is explicit that this factor is a property of the news-silent Robinson–Trautman class, not a fundamental floor: a more collimated exhaust would approach $n^2$6, at the cost of leaving the gravitational-wave–silent class. The cost is severe — reaching $n^2$7 radiates roughly 80% of the rest mass, and $n^2$8 more than 98% — but it is positive and finite.

Existence of admissible accelerating shells

The static anchor is a Schwarzschild–Minkowski thin shell, whose surface DEC holds strictly for compactness $n^2$9 (the flat-interior thin-shell case of the rigorous static Buchdahl bounds of Andréasson and Horvat–Ilijić). The linearized first-junction (Darmois) map at the anchor is the round-sphere rigidity operator $P_B^\mu$0, boundedly invertible off the three-dimensional $P_B^\mu$1 translation kernel by Cohn–Vossen rigidity — so local solvability is a rigidity fact, not an assumption. Continuity of the worst-observer surface DEC margin then gives, via the implicit function theorem, an unconditional (though perturbative) existence theorem: admissible accelerating shells exist in an open neighborhood of the anchor, with the bulk law forcing the binding slice $P_B^\mu$2.

A companion proposition lifts this from a sequence of per-instant snapshots to a genuinely time-evolved spacetime for slow burns, conditional on the dynamical $P_B^\mu$3 persistence of the anchor solvability — the one hypothesis the dynamical result does not prove analytically, though it is realized numerically and checked by a Newton–Kantorovich existence test.

The reconciliation with the companion boundary-cost result (0/600 admissible smooth shells) is direct: that obstruction lived in the finite-width source–vacuum transition, appearing as Hawking–Ellis Type IV stress in the tail. The distributional junction removes the gradient tail, and the non-vacuum positive-radiation exterior removes the "matter dies into vacuum" spike.

The acceleration–compactness frontier

A Buchdahl-type bound $P_B^\mu$4 caps the proper acceleration of a positive-energy shell. The frontier is bracketed:

Quantity	Status	Value
$P_B^\mu$5	numerically verified lower bound	$P_B^\mu$6–$P_B^\mu$7 at $P_B^\mu$8–$P_B^\mu$9, falling to $P_B^\mu$0 at $P_B^\mu$1
$P_B^\mu$2	true nonlinear frontier	no closed form; set by global $P_B^\mu$3 shape response
$P_B^\mu$4	rigorous kinematic ceiling	exact, shape-independent

The rigorous ceiling follows from the exact rear-pole identity $P_B^\mu$5: the areal foliation stays regular only if the rear pole remains outside its acceleration-induced effective horizon, i.e. $P_B^\mu$6. This is the one rigorous closed-form member of the hierarchy; the tighter envelope $P_B^\mu$7 is conceded to be heuristic. Notably, the frozen-shape rear-pole margin actually opens at low order, so the frontier is forced by the global nonlinear shape response — an $P_B^\mu$8 effect, not the linear dipole. A finite-duration bump maneuver at $P_B^\mu$9, $\ell=1$0 is certified admissible at every retarded step, with Bondi balance closing to $\ell=1$1 and a rapidity gain of 0.24 radiating about 51% of the rest mass.

Thick walls, back-reaction, and stability

The thin shell is realized as the limit of finite-thickness admissible matter. The matter model matters decisively: a generic smoothstep Kerr–Schild thickening (radial tension) reproduces the boundary-cost DEC violation in the static limit, whereas a tangential-pressure wall ($\ell=1$2, load carried by $\ell=1$3) is strictly DEC-admissible across its whole width for local compactness $\ell=1$4. The admissibility windows nest by matter model — $\ell=1$5 (collisionless Vlasov, photon-sphere limited) $\ell=1$6 (anisotropic elastic) $\ell=1$7 (distributional shell). Grafted with outgoing null radiation, the idealized $\ell=1$8 two-component stress stays Type I (discriminant $\ell=1$9), and the adiabatic self-consistent Eddington–Finkelstein wall remains a machine-precision Einstein solution ($P_B^\mu = m(u)\,v^\mu(u)$0) with strictly positive margin through $P_B^\mu = m(u)\,v^\mu(u)$1 at tested amplitudes. A substantive sign reversal is reported here: the frozen-background calculation suggests radiation raises the margin, but with the metric free to respond the flux tightens it, $P_B^\mu = m(u)\,v^\mu(u)$2, while preserving admissibility.

On stability, the realized wall sits exactly on the Poisson–Visser marginal curve $P_B^\mu = m(u)\,v^\mu(u)$3 for all $P_B^\mu = m(u)\,v^\mu(u)$4 — a neutral zero mode. Under the frozen-background rear-pole redshift it e-folds in $P_B^\mu = m(u)\,v^\mu(u)$5–$P_B^\mu = m(u)\,v^\mu(u)$6 light-crossing times, and the burn outruns this only in a restricted high-compactness, brief-rapidity corner; the certified $P_B^\mu = m(u)\,v^\mu(u)$7 burn lasts roughly twice the square-pulse lower bound, so even that corner is not certified safe without active stabilization. A slightly stiffer wall is strictly stable at no energy-condition cost (the surface DEC margin is junction-fixed and independent of the equation-of-state slope), so strict stability and strict DEC decouple. The fully dynamical, flux-coupled stability of the radiating shell is left open.

Optimal control and observational signature

Within the gravitational-wave–silent class, the von der Gönna–Kramer converse (vanishing news forces $P_B^\mu = m(u)\,v^\mu(u)$8 emission) makes the Damour dipole the unique silent steering, and a Pontryagin argument shows the minimum-fuel optimum saturates the control law pointwise, $P_B^\mu = m(u)\,v^\mu(u)$9, with minimum $P_B^\mu = m v^\mu$0. The numerical optimizer exhibits the $P_B^\mu = m v^\mu$1 collapse of higher multipoles under any news penalty; the bare ($P_B^\mu = m v^\mu$2) program instead drives a singular rear-pole spike, excluded precisely because it lies outside the silent class. The free-profile problem — optimizing the shape of $P_B^\mu = m v^\mu$3 itself over multi-burn trajectories against the moving frontier — is explicitly deferred.

The mandatory exhaust is an observable: the certified burn peaks at luminosity $P_B^\mu = m v^\mu$4 per light-crossing time, with a single rear lobe vanishing at the forward pole at saturation. No specific photon spectrum is claimed; the construction fixes the energy–momentum budget and angular pattern, not the exhaust microphysics.

Limitations and scope

The paper states its boundaries plainly. The construction is subluminal and causal; it is a rocket, not a reactionless drive; passengers feel genuine proper acceleration (though the flat cavity's tidal tensor vanishes identically, verified to machine precision for $P_B^\mu = m v^\mu$5); the energy cost is astronomical; and the solution exists only on a finite retarded-time slab, with no claim of eternal existence. The existence theorem is unconditional but perturbative, with finite-amplitude reach established numerically (Darmois match to $P_B^\mu = m v^\mu$6–$P_B^\mu = m v^\mu$7, Kantorovich contraction below threshold by $P_B^\mu = m v^\mu$8–$P_B^\mu = m v^\mu$9 at low compactness); a rigorous validated-interval version of the frontier bracket is left open, as are the saturating-amplitude all-angle dynamical certificate, the $\sim 10^{-15}$0 back-reaction (non-negligible at the certified burn's $\sim 10^{-15}$1), and the fully flux-coupled stability analysis. The Santiago–Schuster–Visser no-go is evaded, not contradicted — its hypotheses (vacuum shift-bubble exterior) simply do not describe this construction — and the Ford–Roman quantum-inequality obstruction is inapplicable because no negative energy appears anywhere.

Conclusion

The paper establishes that steering a warp drive is an energy-budget problem rather than an exotic-matter problem: no confined positive-energy drive can change its Bondi four-momentum without radiating, and the photon-rocket warpshell is an explicit positive-energy object that saturates that law, with a closed-form steering law $\sim 10^{-15}$2, an admissible matched shell certified perturbatively and at finite amplitude, a Buchdahl-type acceleration frontier bracketed between $\sim 10^{-15}$3 and $\sim 10^{-15}$4, and a minimum-radiation maneuver that is the Damour dipole. The remaining open questions are concrete: the saturating-amplitude dynamical certificate, rigorous validated bounds on the frontier, the $\sim 10^{-15}$5 back-reaction, flux-couled stability, and multi-burn trajectory optimization over the admissible radiation cone.