Lunokhod 1

NASA/GSFC/Arizona State University. Public domain (NASA / US government work).

Overview

Luna 17, spacecraft Ye-8 no. 203 at 5,700 kg, was launched from NIIP-5 site 81/23 on 10 November 1970 at 14:44:01 UT on a Proton-K with Blok D [3]. After two mid-course corrections it entered an 85 by 141 km lunar orbit inclined 141 degrees and landed at 03:46:50 UT on 17 November 1970 at 38 deg 24 min N, 34 deg 47 min W, in a crater-like depression 150 to 200 m across and 7 m deep, about 2,500 km from the Luna 16 site. Lunokhod 1 rolled down the descent stage’s twin ramps at 06:27:07 UT the same day, taking 20 seconds to reach the surface [3]. The undercarriage was designed and built by VNII Transmash with Lavochkin retaining overall vehicle design.

The last communications session was at 13:05 UT on 14 September 1971; recovery attempts ceased on 4 October [3]. Design life was three lunar days, about 21 Earth days; the rover operated eleven.

Specifications

Parameter	Value	Source
Rover mass	756 kg	[2], [3]
Height	about 1.35 m	[3]
Width across	2.15 m
Length and width	about 2.2 m each	[1]
Wheels	8, independently driven	[3]
Wheel diameter and width	510 mm, 200 mm	[1], [2]
Track gauge	1600 to 1700 mm	[2]
Ground pressure at 30 mm sinkage	0.05 kg/cm2
Speeds available	0.93 km/h and about 2 km/h, two forward and two reverse	[1], [3]
Turning radius, in place	0.8 m	[1], [2]
Solar array capacity	200 A-h, silicon cells	[2]
Night survival	polonium-210 isotopic heat source, lid closed	[1]
Internal temperature regulated to	+10 to +30 degrees C at 735 to 770 mm Hg
Maximum uncommanded motion	27 s per command, 2.5 minute command timeout
Design life	three lunar days, about 21 Earth days	[3]

Mission profile

Parameter	Value	Source
Landing site	38 deg 24 min N, 34 deg 47 min W, Mare Imbrium	[3]
Landing and rover egress	17 November 1970
Last communications session	14 September 1971
Operating life	322 Earth days, 11 lunar days against a three-day design life
Distance driven	10.47 km, revised to 9.93 km in 2013
Mission average speed	about 140 m/h	[2]
Television images returned	more than 20,000	[3]
High-resolution panoramas	206
Soil analyses (RIF-MA)	25
Penetrometer insertions	537, about one per 20 m
Distance per lunar day, days one to six	215, 1504, 1936, 1563, 2004 and 1029 m	[1]

Configuration

The body is a magnesium alloy tub with a convex lid enclosing a pressurized, thermally controlled electronics bay. The lid opens during the lunar day to expose a solar array and closes at night, when a polonium-210 heat source maintains internal temperature [1]. Lunar surface temperature before sunrise is a mean 96 K at the equator and a mean 89 K at 45 degrees latitude [12].

Mobility

Eight wheels, four per side, are each driven by an independent motor, so individual failures do not end mobility [1], [3]. Each wheel is three titanium rings of 510 mm diameter carrying 16 spokes wrapped in metallic mesh, 200 mm wide, on a track gauge of 1600 to 1700 mm; grousers on the wheels disturb the soil enough that the tracks remain visible in Lunar Reconnaissance Orbiter Narrow Angle Camera frames, unlike the ungrousered Apollo LRV wheels at similar wheel loading [2]. Average specific ground pressure at a nominal 30 mm sinkage is 0.05 kg/cm2. Regolith bulk density in the first 30 cm is 1.53 to 1.63 g/cm3, with friction angle 30 to 50 degrees and cohesion up to 1 kPa, and mean particle size 40 to 800 micrometers with the majority at 60 to 80 micrometers [5].

Each wheel hub contains its electric motor, reduction gear, brake, a power-drive disconnection mechanism, a revolution counter, and a temperature gauge [1]. Any wheel can be explosively jettisoned if it seizes, and the vehicle can continue with only two powered wheels per side. Braking is either electrodynamic or by electromagnetic disc brake. Steering is skid steering, by varying the speed and direction of rotation of the appropriate wheels; the same chassis turns in place within a 0.8 m radius by counter- rotating the two sides, and turns while driving on a typical 3 m radius [2].

Two forward and two reverse speeds are available; first speed is 0.93 km/h and second roughly 2 km/h, the latter used infrequently [1], [3]. Because motion is issued in short commanded bursts separated by stops, achieved average speed is 0.01 to 0.17 km/h, and the mission average was about 140 m/h [2], against a nominal top speed of about 100 m/h. The distance considered safe to travel uncommanded is 1 m, half the vehicle length. In the worst observed terrain the wheels sank nearly 20 cm while descending into and climbing out of a boulder-strewn crater at roll angles reaching 30 degrees.

Power and energy

The solar array is the primary source. Its silicon photoelectric cells are mounted on the inner face of the compartment lid, which can be positioned anywhere from 0 to 180 degrees to maximize collection from a low Sun [1], [2]. Total array capacity is 200 A-h, against 250 A-h for the gallium arsenide arrays fitted to Lunokhod 2. Solar flux at the Moon is 1367 W/m2 mean, with a 1315 to 1421 W/m2 range [12]. The array charges a chemical storage battery during the lunar day, disconnecting once the battery is full. Before each night the vehicle is oriented facing east so that the first rays of sunrise fall on the array. Voltage figures are not published; total electric consumption while in motion is described only by comparison to a domestic electric iron, with reserve sufficient that the vehicle could tow an identical vehicle without loss of speed.

The chassis carries the automation equipment, the override safety system, a sensor system for determining soil physical properties and evaluating cross-country performance, the isotopic heat source, and a trailing ninth wheel with its own retraction mechanism [1]. The ninth wheel is undriven and acts as an odometer: comparing its rotation against the driven wheels measures slip directly, and during a difficult crater exit the driven wheels turned while the ninth wheel stood still, the signature of complete loss of traction. An odometer and speedometer are listed among the seven scientific instruments [3].

Thermal

Two circuits, hot and cold, switched automatically by thermal sensors [1]. The cold circuit centers on a radiator in the upper part of the instrument compartment, its external surface nearly ideally smooth and coated with a layer of very low absorptivity and high emissivity, so it radiates without absorbing and continues to reject heat even in direct sunlight. The body carries a thick insulating jacket, but insulation alone is insufficient against solar input, so a continuously running fan circulates a gas mixture through the compartment and delivers excess heat to the radiator. The cold system switches on above 20 degrees C [1].

At night the solar-array lid closes over the radiator, which stops radiating and instead reflects heat back inside. Baffles automatically divert the gas coolant into the heating circuit, where an isotopic source, polonium-210, warms it [1]. The same hot circuit is used at sunrise and sunset, when the array is exposed but incoming flux does not yet offset radiated losses. The temperature regulation system runs autonomously through the night.

Normal internal temperature is +10 to +30 degrees C at 735 to 770 mm Hg [1]. External parts such as wheels and antennas range from -90 to +150 degrees C. During the fourth lunar noon the thermal system withstood 100 degrees C with many external parts above that, while internal temperature stayed within limits. Thermal tests late in a lunar day in January 1971 [1] measured heat loss with the cover open and closed, and on that basis the vehicle was judged safe to ride out the February eclipse with the cover open and the heating system on; it operated normally afterwards despite the array lid reaching +140 degrees C.

Compute and avionics

Lunokhod 1 had no onboard computer in the modern sense. What it had was a logic and memory network that turned six discrete radio commands into timed actuator sequences, plus an independent safety override.

Vehicle movement is controlled entirely by five “go” commands and one “stop” command [1]. These are executed by logic and memory circuits that select a movement mode, store which mode preceded it, and drive the control elements of the wheel motors. The modes are straight-line motion for a programmed time interval and turning for a programmed number of degrees. The rover therefore did not follow a path; it executed a duration or an angle and then stopped. Distance was a consequence of commanded time, which is why the operators reasoned in seconds rather than meters.

The safety override system stops the vehicle automatically when roll or pitch angle limits or wheel motor load limits are reached [1]. Lunokhod 2 extended the same scheme with automatic locks reacting to dangerous slippage on steep slopes and to excess onboard bus voltage [2]. Power can be disconnected from one or several wheels without losing necessary traction. A sensor set continuously measures roll and pitch, traction motor currents, and wheel RPM and temperature. Computing elements work alongside the telemetry system to extract command signals from a high noise background and to handle the incoming data rate, which is of the order of several thousand commands.

Navigation is likewise not onboard. It uses a course gyroscope and a gyro vertical in conjunction with the telephotometers, simultaneous images of the Sun and Earth, and horizon images from the television system [1]. Camera count and placement were chosen to give full view of the surroundings including the horizon, the Sun and the Earth, as well as ground surface information for safe movement [2]. To fix bearing, the vehicle is driven to a flat site and oriented so the telephotometers capture Earth and Sun at once; the data are then reduced by ground computer. Accumulated error by this method over a month of activity was a little over 1 degree [1].

Autonomy

Autonomy in the modern sense is absent. The vehicle has no path planning, no hazard detection, and no onboard perception; it executes timed or angular motion primitives and nothing more. What it does have is a reflexive protective layer that acts without ground involvement: the roll, pitch, and motor-load override that halts motion at limits [1], the 2.5 minute command timeout that halts it in the absence of instruction, the 27 second cap on any single commanded motion, and the thermal system, which switches circuits on sensor thresholds and regulates the compartment autonomously through the entire lunar night when no command link exists at all. The design principle is that the vehicle’s independent behavior is exclusively conservative: everything it does on its own stops something or preserves something.

Communications

The telemetry system uses two frequency bands simultaneously [1]. Command reception is acknowledged, the command executed, and completion signaled back, so the ground sees both receipt and execution as separate events. Round-trip signal time is 2.6 s, and the effective control delay including operator lever actuation about 4.1 s [3].

The antenna complement reflects the split between high-rate imagery and everything else. A steerable high-gain antenna with its own electric drive carries television images. A wide-angle antenna receives radio commands and transmits telemetry. Four whip antennas, mounted two per side, provide additional command reception [1].

Telemetry falls into three categories by phase [1]. Through both day and night the vehicle reports component temperature and pressure, with systems interrogated periodically during the night. While moving it returns wheel temperature and RPM, wheel motor currents, wheel-ground interaction dynamics, television and telephotometer output, and roll and pitch, which together consume a considerable fraction of the downlink. During stops within a duty cycle it returns scientific experiment data.

Optical systems

Two independent television systems serve different purposes [1].

The driving system is two low-resolution small-frame television cameras, designated MKTV, mounted at a height of 950 mm above the surface, the average eye height of a seated person [2], [3]. Frame interval is 3 to 20 seconds depending on terrain and vehicle speed [1], with 20 seconds the routine value. The operator commits to a maneuver and sees the outcome one or more frames later. Lunokhod 2 added a third MKTV camera and, on a higher data rate, reduced the interval to about 6 seconds [2].

The second system is stereo-panoramic, four identical mechanically scanning telephotometer cameras mounted in pairs on each side [1], [3]. Within each pair, one camera scans slightly over 180 degrees horizontally by 30 degrees vertically, the other 360 degrees vertically by 30 degrees horizontally, and one of the two carries a device for determining the local vertical. A panoramic image is about 3000 lines over 180 degrees by 500 samples [2]. Output is telemetered to the ground and reduced by computer into a graphic display on paper tape. Panoramas taken from two rover positions were combined stereoscopically into topographic maps of selected sites, a methodology first developed on Lunokhod 1. The panoramas also image sectors of sky for navigation.

Modes of operation

Operation is confined to the lunar day, in duty cycles, with the vehicle sealed and hibernating through the night [1]. Eleven lunar days were worked over 322 Earth days against a three-day design life [3]. Within a duty cycle the vehicle alternates between commanded motion and stops, with science returned during the stops.

A duty cycle averaged slightly under 5 hours, with about 18 cycles per lunar day, giving roughly 6.5 hours of active driver control across a 14 Earth-day lunar day [1]. Opening the lid to expose the solar array is among the first operations of each day’s work.

Motion itself has an enforced time limit. A single commanded motion may not exceed 27 seconds, so that the vehicle does not overshoot the intended distance [1]. The vehicle is also programmed to stop in place 2.5 minutes after the last command, which is what makes an operator handover safe: the crew simply stops issuing commands and the rover halts itself. Two speeds are available; from the reported 7 minutes 27 seconds to cover a stated distance, first speed corresponds to 0.93 km/hr [1].

Automatic transitions are triggered by the safety override on roll, pitch, or motor load limits [1], and by the thermal sensors switching between hot and cold circuits at 20 degrees C.

Ground operations

Round-trip light time is about 2.6 s, short enough for direct teleoperation, and the ground crew is five people: crew chief, driver, navigator, antenna operator and engineer [1], a composition also recorded as commander, driver, flight engineer, navigator and narrow-beam antenna guidance operator [3].

The crew chief holds overall leadership and all operational decisions, sits behind the driver, continuously verifies the driver’s spoken assessment of the terrain, and alerts the rest of the crew to specific problems [1]. The driver controls motion from a dedicated panel, taking input from a television screen, a light display carrying the basic vehicle telemetry, and information passed orally by other crew members. The navigator performs the navigational computations and produces recommended course data. The antenna operator points the high-gain antenna, and because the television link runs through it, that operator directly determines image quality. The engineer heads a group of specialists performing operational analysis of vehicle-condition telemetry.

Two crews of five, for two shifts, were selected from a pool of hundreds in a process begun in May 1968 [3]; the commanders were Yu. F. Vasil’yev and I. L. Fedorov and the drivers N. M. Yeremenko and V. G. Dovgan’ [3]. Members were aged 28 to 34 and are reported to have received the equivalent of over two years of cosmonaut training before the Luna 17 launch [1]. Teams worked independently and changed frequently within a duty cycle, except that both navigators worked together throughout. Changeover was covered by the 2.5 minute command timeout: the vehicle stops on its own [1]. The effective control delay is about 4.1 s, the 2.6 s round-trip signal time plus the time to move the control levers [1].

The division of sensory channels was deliberate and was itself a subject of study. The driver watches the television image and the roll, pitch, and course needle indicators. Telemetry from internal vehicle systems bypasses the driver entirely and goes to the engineer, who announces it over a loudspeaker [1]. On easy terrain the driver works visually; as terrain difficulty rises the auditory channel takes more load, with the engineer continuously repeating roll and pitch and the crew chief cross-checking terrain assessments. The television image is the only element giving the crew any directness of perception, and the driver integrates the incoming image with remembered images no longer visible in order to maneuver among obstacles.

The driving console is specific. The screen carries a grid overlay: two lines converging on the horizon, marking the track the wheels will follow, crossed by parallel lines serving as a dimensional reference for judging obstacle size [1]. Cameras are mounted at about the height of a standing man’s waist, so the forward view corresponds to a seated eye height. A representative exchange runs [1]: the driver judges the vehicle can safely advance 6 to 7 m; the crew chief orders “first speed forward”; the driver advances the control lever one division forward and presses a button on top of the lever with his thumb; the display returns the word “forward” and the number “1”, while his left hand rests on the brake override button. Returning the lever to neutral displays “stop”, joined after a delay by a second “stop” confirming execution. A turn is commanded by moving the lever to a preset-turn detent, for example PT-20 for a 20 degree turn [1]. A duplicate control system of two rows of buttons without a lever exists, but the lever is preferred because it is less error-prone and gives more feel for the incline.

Driver training targeted the dominant failure mode directly. Errors in estimating the size and distance of obstacles were identified as the main cause of dangerous situations, and a training program built on statistical analysis of those errors reduced them substantially [1]. Antenna operators, crew chiefs, navigators, and engineers were separately trained.

By May 1971, six lunar days of activity had been completed [1]. Excluding the first, which began mid-day, each lunar day of activity ran 12 to 14 Earth days and comprised between 10 and more than 20 duty cycles [1]. Distance per lunar day was 215 m, 1504 m, 1936 m, 1563 m, 2004 m and 1029 m for days one through six, cumulating to 8261 m. The mission total was reported at the time as a 9200 m closed-loop traverse over six months [6] and finally as 10.47 km, revised to 9.93 km on reanalysis in 2013 [3].

Payload

PrOP penetrometer. A conical vane assembly is driven into the soil and rotated, with sensors measuring the force required to rotate it, the rotation angle and the penetration depth [1], [3]. It operates only with the vehicle at rest, and was run every 15 to 30 m of path; 537 insertions were made over the traverse, about one per 20 m.

Instrument parameter	Value
Cone base diameter	5.0 cm
Cone height	4.4 cm
Cone angle	60 degrees
Vane diameter	7.0 cm
Penetration effort	in excess of 20 kg
Vane turn angle	up to 90 degrees

Values from [6].

Bearing strength varied from 0.2 to 1.0 kg/cm2 with terrain condition and rotational shear strength from 0.02 to 0.09 kg/cm2, with 0.34 kg/cm2 and 0.048 kg/cm2 the most frequently occurring values; the soil on the traverse is fine-grained with some cohesion, at least 50 to 100 m thick and uniform in depth, with a friable dust-like top layer, and its mechanical strength increases with depth [6]. Peak penetrometer force was limited by vehicle weight: Lunokhod’s lunar weight is in excess of 900 N, so a single penetrometer could draw only about 39 N of reaction without lifting the vehicle.

What the penetrometer is measuring against is a soil that stiffens quickly with depth. Bulk density is about 1.30 g/cm3 at the surface, rises to 1.52 g/cm3 at 10 cm and 1.83 g/cm3 at 100 cm, and approaches 1.92 g/cm3 asymptotically below that, following a hyperbolic fit in depth [8]. Micrometeorite gardening loosens only the top 10 to 15 cm; below it lies a layer the same flux has packed tight, and porosity falls from about 65 percent at the surface to under 40 percent at depth.

The ninth wheel provides the complementary in-motion measurement, rolling without slipping so that its count against the driven wheels yields traction and slip [1].

RIF-MA, an X-ray fluorescence spectrometer [3]. An isotopic source irradiates a sector of soil, inducing characteristic X-ray emission from its constituent elements. The receiver is a set of purpose-built proportional counters with characteristic filters, read out by a 64-channel amplitude analyzer [1]. The unit hangs between the front wheels, and drivers were cautioned to avoid stones large enough to damage it.

X-ray telescope. Two proportional counters covering 2,000 to 10,000 eV, or 1 to 6 angstroms [1], behind collimators restricting the field of view to a 3.5 degree cone, both pointed at the local zenith when the vehicle is level. One counter is covered by a filter opaque to the band of interest and so records background only, allowing the other’s count to be differenced against it [1]. Observations are made only while stationary, the pointing sweeping 9 to 10 degrees over 18 to 20 hours as the Moon rotates. During the second lunar day, at least 4 discrete sources were identified among 33 sectors studied [1].

Radiometric equipment, measuring corpuscular cosmic ray fluxes in energy regions inaccessible from the Earth’s surface [1].

TL-1 laser retroreflector. A 3.7 kg French-supplied panel of 14 four-sided silica glass prisms of 10 cm, mounted above the forward cameras, each prism reflecting an incoming beam three times so that it returns along its arrival path regardless of incidence angle [1], [3]. The prisms are of unusually uniform glass with low thermal expansion behind a thermal shield, built by Sud Aviation. Ranging used a Q-switched ruby laser from the Lebedev Physics Institute on the 2.6 m telescope at the Crimean Astrophysical Observatory [1], firing a pulse every 15 seconds during a session into a narrow-band photoreceiver, with accuracy of a few meters at the time and an expectation of reaching 10 to 20 cm. First use was on 5 and 6 December 1970, giving an Earth-Moon distance to 30 cm [1]; dust apparently covered the reflector and few further echoes were obtained. Ranging resumed only after Lunar Reconnaissance Orbiter imagery recovered the rover’s position [2].

Technologies developed

The chassis and sealed instrument compartment were reused unchanged on Lunokhod 2, which massed 836 kg and improved the array to 250 A-h gallium arsenide, added a third navigation camera, and added automatic locks reacting to dangerous slippage on steep slopes and to excess bus voltage [2].

Penetrometer data from the two Lunokhods is the larger of the two lunar in-situ geotechnical datasets and remains a basis for the depth dependence of lunar soil strength; the measurements were summarized in the Lunar Sourcebook, where the cone data are inverted for cohesion and friction angle at several terrain types [12], and the depth dependence they constrain is what footpad, instrument deployment and wheel-soil interaction analyses rest on [8]. No true Bekker parameters exist for lunar regolith, since that data would have had to be collected in situ; the Apollo-era parameters were estimated from Surveyor data, should be treated as rough estimates only, and are themselves dependent on wheel size and loading conditions [4].

The laser retroreflector remains operational. Reflector coordinates established to centimeter level define geodetic reference points for the lunar coordinate systems in current use [2].