Dropped from a drone, this model rocket nearly lands on its motor

Oct 11, 2026 Last reply: 2 hours ago 2 Replies
Jump to replies (2)

Landing a model rocket under power is one of the hardest things you can ask of a small airframe: it has to fall, light its motor at the right moment, and touch down upright. The maker's Echo rocket exists to practice exactly that, and the latest attempt came very close to a clean propulsive landing, falling just short at touchdown.



The build itself is a tidy package. Echo stands about one meter tall and weighs one kilogram, finished in dark blue or black with white markings on the side. It flies on an Estes F15 solid motor, steers with a thrust vector control mount, and runs a Signal flight computer. The landing legs are 3D printed from PLA. For this test, the fourth landing attempt, the rocket was released from a drone at 30 meters, then ignited its motor to slow down and settle upright.



The details that stand out to me are in the sensing work. The maker put real effort into sensor calibration, and orientation estimation moved from accelerometer-based to gyroscope-based using quaternions. On-screen graphics tracked altitude, velocity, and orientation rates through the landing burn, and the footage is set against earlier attempts, which makes the progress easy to follow.



There are honest limits, though. Leg deployment caused vibration that disturbed the gyroscope readings, and that led to a slight tilt on touchdown. Solid motors offer no throttle control, which the maker calls out as a real challenge. Planned fixes include better data filtering and retiming ignition for a higher hover, and further down the line the maker wants a single vehicle, called Scout, that both launches and lands. BPS.space is clearly iterating methodically, and this test looks like a strong step toward the goal.



Project by BPS.space. Watch it on their YouTube channel:

formatting link


The orientation work sounds like the key piece here. I am curious how the maker tracked attitude during the descent, since a short drop from a drone leaves very little time for the sensing to settle before the burn starts.

From what the maker shared, orientation sensing started out accelerometer-based and transitioned to gyroscope-based using quaternions, with dedicated calibration work done beforehand for the landing. It is also worth noting the legs caused vibration during deployment that disturbed the gyroscope readings and produced a slight tilt on touchdown, and better data filtering is listed among the planned improvements.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required