The Elevated Evacuated Tube (EET) is a low-air-resistance corridor through which spacecraft fly to avoid aerodynamic drag and heating in the densest part of the atmosphere. Unlike the horizontal acceleration section and the ramp section, the EET is not designed to accelerate the spacecraft or redirect it skywards; therefore, the spacecraft is neither connected to nor guided by the EET. Instead, the EET is positioned precisely around the spacecraft's flight path.

The primary control objective of the EET is to remain centered on the predicted flight path, which serves as the common reference trajectory for both the vehicle and the guideway. Deviations from this objective can be decomposed into two error classes: flight-path tracking error, defined as the difference between the spacecraft’s actual trajectory and the predicted trajectory; and EET navigation error, defined as the difference between the EET centerline and the predicted trajectory. The allowable tube clearance must accommodate the superposition of these error sources under worst-case disturbances.

Flight Path Tracking Error

The guideway in the ramp is engineered with millimeter-scale accuracy, as was discussed in the claim The Required Geometric Straightness of the Underground Ramp Section is Achievable. The spacecraft’s speed can be fine-tuned by imparting a small excess velocity in the horizontal acceleration section and then bleeding off that excess in the ramp section using dynamic eddy-current braking, while continuously and precisely measuring the spacecraft’s speed. After the spacecraft decouples from the launch sled at the end of the ramp section, it will be travelling on a hyperbolic trajectory while still in a vacuum.

From this point on, inertia takes over as the spacecraft travels through the 102-km-long EET for roughly 9 seconds. As there will be a small amount of residual air in the elevated evacuated tube, the spacecraft can orient itself using its reaction control wheels so that the air flowing past the spacecraft is harnessed to help keep it travelling along the predicted flightpath.

In practice, algorithms that predict the path of a spacecraft must account for the complex gravitational influence of the Earth, the Moon, and the sun. The Earth's gravitational field has already been accurately mapped. Algorithms will account for the effect of tides, and the nature of the terrain near that launch system. The current state of the art is perhaps best illustrated by the twin, co-orbiting GRACE (Gravity Recovery and Climate Experiment) satellites. These satellites, launched in March 2002, achieved an absolute positional accuracy of better than 5 cm in each direction (ref).

Other forces typically considered for orbit determination include the solar radiation pressure (which will be zero because the spacecraft is inside an aluminum tube) and outgassing (which will be effectively eliminated through good spacecraft design and preparation). Consequently, the only significant force on the spacecraft is its interaction with the residual air in the evacuated tube, and this force is actively harnessed to prevent the spacecraft from drifting off the predicted flightpath. For these reasons, it is anticipated that cm-level alignment between the actual flight path and the predicted flight path can be achieved in practice.

EET Navigation Error

If the predicted and actual flight paths are closely aligned, then the remaining challenge is positioning the EET around the predicted flightpath with sufficient accuracy. Drone light shows, which precisely position hundreds of drones to create impressive arial artwork, illustrate how precisely autonomous flying machines are already able to position themselves in the sky. In fact, on October 17th, 2025, a light show comprising 15,946 drones was held over Liuyang City in central China's Hunan Province setting a new world record (video link). The EET would require

$$N_{fans} = EET_{Length} \times FansPerMeter = 102,040 \times 2/18 = 11,338 \space fans$$

So this record setting drone light show involved more drones than there are lift fans on the EET in the digital twin reference design.

Similar technologies include the U.S. Navy's automated and precision-assisted landing technologies such as the Joint Precision Approach and Landing System (JPALS), which helps to precisely land aircraft on carriers to reduce wear on aircraft landing gear and deck equipment by smoothing approaches and reducing hard landings. In fact, these systems have demonstrated such high repeatability and positional accuracy that operational procedures intentionally introduce controlled variation in touchdown location to avoid concentrating wear on a single region of the carrier deck.

These examples demonstrate that precise, closed-loop positioning relative to a moving or predicted reference is already a solved problem at comparable or greater levels of complexity. In both drone light shows and automated carrier landings, distributed autonomous systems use sensor fusion, state estimation, and feedback control to continuously minimize deviation from a time-varying reference trajectory under uncertainty and disturbances. The EET navigation problem is structurally similar: rather than controlling a vehicle to a fixed runway or formation pattern, the task is to control a guideway’s centerline relative to a predicted flight path. Given that existing systems routinely achieve centimeter- to decimeter-scale tracking accuracy over large spatial scales, it is reasonable to conclude that an actively controlled EET can be made to center itself around a predicted trajectory with the precision required to materially reduce tube diameter and associated costs.