Colmena

LINX, Instituto de Ciencias Nucleares, UNAM. Used with attribution on the rights holder’s terms.

Overview

Colmena is a swarm of five micro-rovers built by the Laboratory of Space Instrumentation (LINX) at the Institute of Nuclear Sciences of the National Autonomous University of Mexico, flown as an integrated payload on Astrobotic’s Peregrine Mission One [8]. Each rover is a 12 cm disc weighing about 56 g. The five together weigh 280 g, and the whole payload, including the module that stores, deploys and talks to them, is 608 g.

The name is Spanish for beehive, and the architecture follows from that analogy. The project’s stated end is mineral prospection: lunar surface resources are scattered over large areas rather than concentrated in deposits, and the response is to distribute the work across many small, simple units that survey individually and act collectively [8]. Regolith depth itself varies by an order of magnitude across the near side, from a few meters to tens of meters, by both radar and crater-count surveys [4][6], which is the kind of spatial variability a scattered, individually surveying swarm is suited to sampling. Colmena-1 is the first of at least three missions [8][13]. It was a proof of concept for miniaturization and a probe of the lunar surface limit layer; Colmena-2, planned for 2028, is to deploy at least three prospection rover prototypes with a positioning system, beyond-the-horizon navigation and one night in hibernation; Colmena-3, planned for 2031, adds sample collection and transport and survival across several diurnal cycles.

Peregrine launched on 8 January 2024 on the first flight of Vulcan Centaur [2]. A helium pressure control valve lost seal capability, most likely through vibration-loosened threads in a joint internal to the valve, and let helium flow uncontrolled into the oxidizer tank until it ruptured about 92 minutes after separation, ruling out a lunar landing [2]. The team recovered attitude control and operated the spacecraft for ten days and fourteen hours in cislunar space before a controlled reentry over the South Pacific on 18 January [2]. Colmena was never deployed.

Specifications

Parameter	Value
Rovers	5, identical
Rover mass	about 56 g each, 280 g total
Rover envelope	approximately 12 x 12 x 4 cm
Central torus thickness	approximately 1 cm
Wheels	2, plus four skates
Solar panels	on both faces
Deployment module mass	328 g
Payload total mass	608 g
Standby temperature range	-180 to 150 C
Operating temperature range	0 to 120 C
Radiation survival requirement	120 days in cislunar space outside the magnetopause
Catapult range	10 to 15 m
Compressive force per wheel	under 5 g

Source: [8]. Speed, drive torque, battery capacity, generation, processor and radio parameters are not published.

Mission profile

Parameter	Value	Source
Lander	Astrobotic Peregrine	[8]
Launch	8 January 2024, Vulcan Centaur maiden flight	[2]
Anomaly	helium pressure control valve loss of seal, oxidizer tank rupture about 92 minutes after separation	[2]
Duration operated	10 days 14 hours	[2]
Distance traveled	about 860,000 km	[2]
Reentry	18 January 2024, South Pacific	[2]
Technical objectives met	75 percent, by the project’s own accounting	[8]

With landing impossible, the payload was turned on in cruise, which it had not been designed to do and which was itself a thermal problem. Every operational mode was exercised, telecommunications were run, telemetry and thermal data were acquired to validate the thermal architecture, and the electronics were checked for single event effects after crossing the Van Allen belts. The project’s own accounting of 75 percent [8] covers design, construction, qualification, launch survival and that cruise operation. It carries no per-objective breakdown, and every objective that required contact with the lunar surface is among the missing quarter.

Mobility

Each rover is a disc with two wheels and four skates, reversible about its mid-plane, with solar panels on both faces so it works whichever way up it lands [8]. That symmetry is not a preference; it is forced by the deployment method, and it costs the design a doubled set of solar panels.

The wheels are not primarily for locomotion. The rovers were designed to interact with the regolith in the ways their science required rather than to travel: measuring the resistance of the regolith to wheel movement at different torques is itself one of the measurements [8].

The reason the mobility problem is different at this scale is quantified. A heavy rover compresses regolith until the irregular larger grains mechanically lock and the ground behaves as a solid [10]. A Colmena rover applies under 5 g of compressive force per wheel, and at that loading the intermediate, micron-scale fraction of the dust may behave as a dry fluid-like upper layer instead [8]. Whether it does is one of the questions the mission existed to answer.

Deployment

The rovers ride inside a bucket under a removable shroud, mounted on the arms of a catapult in the Telemetry, Telecommunications and Deployment Module, which stays bolted to the lander deck for the whole mission [8]. The catapult is armed by hand during payload integration and holds that mechanical energy until a cutting system, commanded from the ground through the module’s own electronics, releases it.

It throws the rovers 10 to 15 m from the lander [8]. The distance is set by two constraints: the rovers must clear the lander’s shadow, where there is no solar power, and they must land outside the area of regolith disturbed by the descent thrusters, which would otherwise impede them [8]. That hazard is documented from Apollo 12: the nearby Surveyor 3 spacecraft, 155 m from the landing site, accumulated significant dust deposited by the Lunar Module’s descent engine [7]. A third property is a benefit rather than a requirement. The catapult distributes the rovers randomly, which is what their collective task needs. The price is that their landing orientation cannot be foreseen, which is where the reversible architecture comes from.

The shroud and bucket also serve as thermal and ultraviolet protection during the cislunar cruise, and the shroud is vented for the rapid decompression inside the launch fairing [8].

Power and energy

Solar panels on both faces of each rover, with a power supply system and an emergency power backup on the central electronics plane [8]. No generation figure, battery capacity or duty cycle is published.

Thermal

The rovers were designed for -180 to 150 C in standby and 0 to 120 C in operation, with the shroud and bucket providing thermal protection until deployment [8]. Validating the thermal architecture was one of the objectives actually met, from telemetry acquired during the unplanned cruise operation, though no temperature values from that operation are published.

Compute and avionics

Each rover carries an onboard computer, power supply and backup, data preprocessing and storage, a telecommunications system, two motors, and its sensors, all on the central plane of the torus [8]. The deployment module’s electronics are on a single flexible printed circuit board beneath the bucket: power regulation and conditioning, onboard computer, data storage, telecommunications and antenna, housekeeping, thermal sensors, and the arm, trigger and cutting subsystems.

Radiation tolerance was approached by design rather than by shielding. At a 608 g payload mass no conventional shielding was affordable, so new approaches in design, materials, integration technique and local shielding were developed instead [8]. Passive survival of the electronics through the Van Allen belts, and their resilience to single event effects in cislunar space, were among the objectives the cruise operation validated. No processor part, memory size or dose figure is published.

Autonomy

Self-organization is the project’s premise, consistent with the case made elsewhere for swarm robotics in space missions: autonomous units with only local sensing and communication, no centralized control, cooperating toward a shared task [3]. The ballistic deployment makes some of it unavoidable here: five rovers scattered at unknown positions and unknown orientations cannot be driven individually from Earth. Each carries its own computer and preprocesses its own data [8]. No navigation method, hazard response or fault behavior is published, and none was exercised.

Communications

The deployment module is the communications hub. It is the interface between the rovers and the lander, and through the lander to the Colmena ground control center, and it preprocesses and compresses both science and engineering data before passing it on [8].

Radio is also an instrument here. The rovers exchange signals at several frequencies among themselves and with the module, and the attenuation of those signals is one of the measurements of the dusty plasma layer.

Payload and instruments

Each rover carries sensors embedded in the torus that sample time-varying ambient magnetic fields between 1 Hz and 100 kHz, three-axis accelerometers, three-axis magnetometers, temperature sensors, and an electromagnetic plasma probing subsystem [8]. The deployment module contributes environment and structural temperature sensing and works cooperatively with the rovers as a limit layer plasma probe.

The target is the layer within roughly 20 to 30 cm of the surface. Ultraviolet photons charge the uppermost regolith positively, producing a vertical electrostatic field [5][11]; photoelectric yield measurements on returned Apollo and Luna dust grains show the effect growing by an order of magnitude from sub-micron to several-micron grain radii before leveling off [1]. Sub-micron grains ejected by that field mix with the solar wind to form a dusty plasma, are observed jumping centimeters above the surface under ultraviolet or plasma exposure [12], and either levitate or move ballistically, with a Debye length near the surface of the order Colmena targets [8][11]. Simulations in the paper show this plasma coupling appreciably to the mechanics, electronics and telecommunications of a device immersed in it, which is exactly where a 12 cm rover sits and where a conventional rover does not.

The planned measurements were the magnetic field fluctuation spectrum from 1 to 10^5 Hz, radio attenuation between rovers, induced potential differences between structural pieces, temperatures very near the surface, and regolith resistance to wheel motion at different torques [8].

Modes of operation

Stowed under the shroud during launch, cruise and landing; deployed by catapult; and operating on the surface as a swarm [8]. Only the stowed mode was flown, and the payload was additionally operated in an unplanned cruise mode outside the magnetosphere, which exercised all of its operational modes electrically without any of them acting on regolith.

Ground operations

The Colmena Ground Control Center commands the payload through the lander’s electrical and logical interface, which also supplies power and carries telemetry [8]. The catapult is fired only after separate arm and trigger commands from the ground.

Technologies developed

The transferable result is a demonstrated design point: a functioning lunar rover, with computer, power, storage, telecommunications, motors and a sensor suite, at 56 g and 12 cm, qualified for launch vibration and shock and for the cislunar radiation environment without conventional shielding [8]. Micro-scale guidance and control components for spacecraft, built from microminiature sensors, actuators and VLSI microcomputers, were proposed as a distinct engineering line as early as 1992 [9]; Colmena-1 is a flown instance at the low end of that scale. The mass budget forced new solutions in design, materials, integration technique and shielding, and those are what the project carries forward into Colmena-2.

The deployment architecture is the second result. Arming a catapult during integration and firing it from the ground removes any deployment mechanism from the rovers themselves, places them clear of both the lander shadow and the thruster-disturbed regolith, and randomizes their distribution, at the cost of a mechanically symmetric rover with twice the solar panels.

More than 250 students across engineering, physics, actuarial science, mathematics, geology and design took part over eight years, and the project treats that as one of its outputs [8].

References