A History of Robotic Arms in Space

The Helping Hand

Robotic arms are among the most essential and recognizable tools in space exploration. They are the remote hands of humanity, allowing astronauts and ground controllers to build, service, and investigate objects in environments far too dangerous or distant for direct human touch. From the cargo bay of the Space Shuttle to the dusty plains of Mars, these manipulators have evolved from simple scoops into highly autonomous, multi-jointed limbs that are foundational to modern spaceflight. They build space stations, repair billion-dollar telescopes, grapple visiting cargo ships, and perform delicate scientific analysis on other worlds.

The need for such technology is driven by the physics of space. Extravehicular Activity (EVA), or a spacewalk, is one of the most hazardous things an astronaut can do. It’s physically exhausting, and the risk of suit punctures or tethers breaking is ever-present. Furthermore, humans are simply not built for the environment. We can’t survive the vacuum, the extreme temperatures, or the radiation. Robotic arms are perfectly suited for it. They are immensely strong, never tire, and can be designed with specialized “hands” and “eyes” for specific tasks. They can also be controlled with superhuman precision, either by an astronaut at a nearby console or by a team of engineers millions of miles away on Earth.

This is the story of their evolution, from the first tentative samplers of the 1960s to the intelligent, self-directed systems of the 21st century.

The Dawn of Space Robotics: Early Concepts and Precursors

Long before the first real hardware was built, the idea of robotic arms in space was a staple of science fiction. But the first practical applications came from the need to answer a simple, significant question: what are other worlds made of? To answer that, you had to touch them.

The Soviet Union’s Luna program pioneered robotic sample return. While not a complex “arm” by modern standards, the Luna 16 lander, which arrived at the Moon in 1970, was equipped with a robotic drill. This drill arm extended from the lander, bored into the lunar surface, and transferred a core sample of regolith into a spherical return capsule. The capsule then blasted off, returning the 101-gram sample to Earth. It was a remarkable achievement, the first time a robotic spacecraft had collected an extraterrestrial sample and brought it home. This was repeated by Luna 20 (1972) and Luna 24 (1976), proving that automated science collection was possible.

The United States had its own early robotic scientist. The Surveyor program was designed to soft-land on the Moon to test the surface for the upcoming Apollo program. Beginning with Surveyor 3 in 1967, the landers carried a “soil mechanics surface sampler.” This was a simple, 4-jointed arm with a small scoop at the end. Controlled from Earth, it was used to dig trenches, poke at rocks, and test the bearing strength of the lunar soil. The Apollo 12 astronauts famously landed near Surveyor 3 two and a half years later. They walked over to the robotic lander and, using hand tools, removed its camera and the robotic arm’s scoop to bring them back to Earth. It was the only time a piece of a robotic probe was retrieved from another world by humans, a symbolic passing of the torch.

Even the hostile surface of Venus was touched by robotic arms. The Soviet Venera program landers, starting with Venera 9 and Venera 10 in 1975, carried simple manipulators. On Venera 13 and Venera 14 (1982), an arm deployed from the lander to press a sensor against the ground, measuring its compressibility. Another arm system was designed to drill into the rock and bring a sample inside the lander for analysis. Given the crushing pressure and 465°C (869°F) temperature, these arms had to work perfectly in a matter of minutes before the lander was destroyed by the environment. These early systems were not versatile, but they were the necessary first step, proving that remote manipulation was a viable tool for planetary science.

The Arm That Built a Legend: The Shuttle’s Canadarm

The true icon of space robotics arrived with the Space Shuttle program. The Shuttle was designed to be a versatile space truck, capable of deploying and retrieving satellites. To do that, it needed a crane. This need gave birth to the Shuttle Remote Manipulator System (SRMS), known to the world as the Canadarm.

Genesis of a National Icon

The arm was a contribution from Canada. In an agreement with NASA, Canada would design, build, and deliver the first flight-qualified arm at no cost. In return, NASA would purchase subsequent arms and, most importantly, Canadian astronauts would get to fly on the Shuttle to operate the hardware they had built. This program was managed by the National Research Council of Canada and built by a consortium of companies led by Spar Aerospace (now part of MDA).

The Canadarm was an engineering marvel. It was 15.2 meters (50 feet) long and about 38 centimeters (15 inches) in diameter. Deceptively lightweight for its size, it was built of advanced carbon composite materials. On Earth, it was so long and delicate that it couldn’t even lift its own weight; its motors weren’t strong enough to counteract gravity. But in the microgravity of orbit, it was a titan, capable of maneuvering payloads the size of a city bus, weighing up to 29,000 kilograms (65,000 pounds).

Its design was modeled on the human arm. It had six degrees of freedom, meaning it had six joints that allowed it to bend and twist in space. It had “shoulder” joints (two) where it attached to the Shuttle’s payload bay, an “elbow” joint (one) in the middle, and three “wrist” joints (pitch, yaw, and roll) at the end. This gave it the flexibility to reach anywhere in the cargo bay or grapple a satellite flying nearby.

The “hand” of the Canadarm was not a gripper. It was a device called an end effector. It looked like a cylinder with three snare wires inside. To grab something, the astronaut would fly the end effector over a special “grapple fixture” on the payload – a simple pin with a target on it. The end effector would slide over the pin, and the astronaut would command the wires to snare, pulling the pin tight and creating a rigid, unshakeable connection.

Controlling this giant limb was a delicate art. An astronaut, typically the mission specialist, would stand at the aft flight deck of the Shuttle, looking out the back windows into the payload bay. They used two hand controllers: one for rotation (pitch, yaw, roll) and one for translation (moving up/down, left/right, forward/back). Cameras on the arm itself – one at the “elbow” and one at the “wrist” – provided the primary views. Because the arm was so large and the payloads so massive, movements had to be slow, gentle, and deliberate. Sudden stops could send oscillations rippling through the arm, a phenomenon known as “wobble.”

A Career of Milestones

The Canadarm made its debut on the second-ever Shuttle flight, STS-2, aboard Space Shuttle Columbia in 1981. It passed its tests with flying colors, moving through its full range of motions. Its first operational use came on STS-3, where it was used to deploy and test a plasma diagnostics package. But its true potential was soon realized in a series of dramatic retrieval and repair missions.

Satellite Deployment: The arm’s bread-and-butter job was deploying satellites. The astronaut would lift a satellite, attached to a “spin table,” out of the bay, point it in the correct direction, and release it. The satellite would then fire its own motor to head to a higher orbit. The arm deployed dozens of satellites this way, including many Tracking and Data Relay Satellites (TDRS) and communications satellites for companies like Intelsat.

Satellite Repair: The real-world value of the arm was proven in 1984 on mission STS-41-C. The Solar Maximum Mission (SolarMax) satellite had failed, and NASA launched a bold mission to repair it in orbit. The plan was for astronaut George “Pinky” Nelson to fly over to the satellite in the Manned Maneuvering Unit (MMU) and attach a docking device so the Canadarm could grab it. But the device failed to latch. The satellite, bumped by the attempt, began to tumble.

Thinking fast, the ground controllers stabilized the satellite, and the Shuttle crew tried a new plan. Shuttle commander Robert Crippen painstakingly flew the orbiter so close to the satellite that the arm operator, Terry Hart, could reach out and grapple it. It was a masterful act of remote-control “fly-catching.” Hart snared the grapple fixture, pulled the 2,300 kg (5,000 lb) satellite into the payload bay, and locked it into a repair cradle. Astronauts Nelson and James “Ox” van Hoften then performed an EVA to replace the faulty electronics module. The Canadarm then lifted the repaired SolarMax back into orbit, a triumphant success that proved the concept of on-orbit servicing.

This was not an isolated event.

• On STS-51-A in 1984, the arm was used to retrieve two stranded communications satellites, Palapa B2 and Westar 6, which had been deployed on a previous mission but failed to reach their proper orbit. The arm snagged them and placed them in the payload bay for return to Earth and refurbishment.
• On STS-49 in 1992, the arm was part of the famously difficult Intelsat VI F-3 satellite rescue. After two separate EVA attempts to attach a grapple bar failed, a new plan was devised. In a historic, three-personEVA, astronauts Pierre Thuot, Rick Hieb, and Thomas Akers positioned themselves around the massive satellite. As commander Daniel Brandenstein held the Shuttle steady, the three astronauts grabbed the 4,215 kg (9,292 lb) satellite with their gloved hands and manually guided it while the arm operator, Richard Hieb, carefully maneuvered the arm’s end effector to a different, unintended grapple point. Once captured, the arm held the satellite while a new kick-motor was attached, saving the mission.

The Hubble Space Telescope

The Canadarm’s most famous partner was the Hubble Space Telescope. The arm was involved in every single NASA mission to the iconic observatory.

Deployment (STS-31, 1990): The arm was responsible for lifting the 11,110 kg (24,500 lb) telescope – the size of a school bus – out of Discovery’s payload bay and gently releasing it into orbit. The release was tense, as one of the solar arrays was slow to unfurl, but the arm held Hubble steady until the problem was resolved.

Servicing Mission 1 (STS-61, 1993): This was perhaps the most high-stakes repair mission in history. Hubble’s primary mirror was flawed, and astronauts had to install corrective optics. The mission began with astronaut Claude Nicollier using the Canadarm to capture the giant telescope and berth it in the payload bay. For five straight days of EVAs, the arm was indispensable. It acted as a “cherry picker” for the spacewalkers, with an astronaut anchored to the end of the arm by a Manipulator Foot Restraint (MFR). The arm operator, usually Story Musgrave, would “fly” the EVA astronaut to different work sites, moving them around the 13-meter-long telescope. The arm was also used as a crane, carefully extracting the old Wide Field and Planetary Camera (WFPC) and inserting the new, corrected Wide Field and Planetary Camera 2 (WFPC2). It then maneuvered the telephone-booth-sized COSTAR instrument into place. The mission was a complete success, and the arm was a primary tool.

Servicing Missions 2, 3A, and 3B: This pattern was repeated on four subsequent missions.

• SM2 (STS-82, 1997): The arm captured Hubble and was used to install the Space Telescope Imaging Spectrograph (STIS) and the Near Infrared Camera and Multi-Object Spectrometer (NICMOS).
• SM3A (STS-103, 1999): An emergency mission to replace failing gyroscopes. The arm captured the telescope and served as the EVA platform for astronauts to swap all six gyros and a new main computer.
• SM3B (STS-109, 2002): The arm again captured Hubble. This time, it supported the installation of the new Advanced Camera for Surveys (ACS) and, in a delicate operation, the replacement of Hubble’s Power Control Unit (PCU), its main power-switching box.

Servicing Mission 4 (STS-125, 2009): The final, most complex servicing mission. The arm captured Hubble one last time. Over five grueling EVAs, astronauts used the arm as a mobile base to install Wide Field Camera 3 (WFC3) and the Cosmic Origins Spectrograph (COS). Most impressively, they performed repairs on instruments never designed to be fixed in space. This involved the arm holding an astronaut steady for hours while they opened access panels and swapped out individual circuit boards. The arm’s precision was essential. At the end of the mission, the Canadarm lifted Hubble from the bay and released it, its “old friend,” for the last time.

The Inspector’s Tool: Post-Columbia Era

The Space Shuttle Columbia disaster in 2003 changed the arm’s job description. The investigation found that the orbiter was fatally damaged by a piece of foam striking its wing during launch. For the “Return to Flight,” NASA mandated that every Shuttle be thoroughly inspected in orbit.

The 15-meter Canadarm couldn’t reach all the spots that needed to be checked, like the orbiter’s belly or the leading edge of its wings. The solution was the Orbiter Boom Sensor System (OBSS). This was a 15-meter (50 ft) long extension boom, essentially a second, simpler “arm” that attached to the end of the Canadarm. This “boom on an arm” effectively doubled the system’s reach to over 30 meters (100 feet).

The OBSS was equipped with a sophisticated package of lasers and cameras at its tip. On every flight after STS-114, astronauts would grapple the OBSS, lift it out of the bay, and spend hours meticulously “scanning” the thermal protection system (the heat shield tiles and wing panels). This data was sent to Earth, where analysts combed through it to find any damage.

This system wasn’t just for inspection; it became a repair tool. On STS-114 itself, two “gap fillers” (ceramic-coated fabric) were seen protruding from the orbiter’s belly tiles. It was feared they could cause dangerous turbulence during reentry. Astronaut Steve Robinson was attached to the end of the Canadarm, which was holding the OBSS, which in turn had Robinson at its end. The arm system carefully maneuvered him under the orbiter, where he simply pulled the fillers out with his hand.

The most dramatic use of the arm-and-boom combination came on STS-120 in 2007, during an ISS assembly mission. A newly deployed solar array on the station’s P6 truss tore, and NASA had to find a way to fix it. The problem was that the tear was at the far end of the truss, 27 meters (90 ft) from the station’s mobile transporter, and the array was still partially energized.

The solution was a spacewalk of incredible complexity. Astronaut Scott Parazynski was placed at the end of the OBSS, which was attached to the Canadarm (which was, in turn, attached to the Shuttle). The Shuttle’s arm operator, Stephanie Wilson, then “flew” Parazynski out to the damaged array. This entire 30-meter structure had to be maneuvered with millimeter-precision to avoid touching the “live” solar array. Parazynski, insulated with homemade “cufflinks” and tools wrapped in tape, successfully stitched the array back together. The arm and boom system was the only platform that had the reach and stability to make the repair possible.

The original Canadarm flew on 90 Shuttle missions before it was retired with the fleet in 2011. The arm from Space Shuttle Endeavour is now displayed at the Canada Aviation and Space Museum in Ottawa.

Building the International Space Station: Canadarm2 and Dextre

The success of the Canadarm led to Canada being invited to provide the next generation of robotics for the International Space Station (ISS). This wasn’t just a single arm; it was an entire suite of tools called the Mobile Servicing System (MSS). Its centerpiece is the Space Station Remote Manipulator System (SSRMS), universally known as Canadarm2.

Canadarm2: The Station’s Workhorse

Canadarm2 arrived at the ISS in April 2001, delivered by STS-100. It is a significant evolution from its predecessor. It’s bigger – 17.6 meters (57.7 ft) long – and much smarter. While the original arm had six joints, Canadarm2 has seven. This extra joint makes it “redundant,” giving it flexibility more like a human arm, allowing it to bend around obstacles.

But its single greatest innovation is that it can move. The original Canadarm was bolted to the Shuttle. Canadarm2 is a “walker.” It has a Latching End Effector (LEE) – the “hand” – on both ends. The station is dotted with Power and Data Grapple Fixtures (PDGFs). The arm can plug one end into a PDGF, which provides it with power, data, and a physical anchor. It can then unplug its other end and “walk,” inchworm-style, to another PDGF, plugging in and establishing a new base.

This allows the arm to roam around the station’s exterior. It can ride along the main truss on the Mobile Base System (MBS), a sort of railway cart. This gives it access to the entire 108-meter (356 ft) Integrated Truss Structure.

Canadarm2 also has a much more advanced brain. It features “force-moment sensing,” allowing it to “feel” resistance. If it’s moving a large object and bumps into something, it can automatically stop before it causes damage. It can also be controlled from multiple locations: by astronauts inside the station, typically from the “Robotic Workstation” in the Cupola module, or by controllers on the ground at NASA’s Johnson Space Centerin Houston or the Canadian Space Agency (CSA)’s headquarters in Quebec.

The Station Builder

Canadarm2’s primary job for its first decade was to build the International Space Station. It was the construction crane that assembled the largest structure ever put in space.

Its assembly career began almost immediately. After its installation, it was used to “hand over” the cradle it launched on to the original Canadarm on the Shuttle, a symbolic “passing of the torch” from one generation to the next.

From there, it became the station’s primary construction tool. As the Space Shuttle arrived with new modules or truss segments, Canadarm2 would take over. The Shuttle’s arm would often lift the new component out of the payload bay, and Canadarm2 would grapple it, taking possession. Then, with astronauts and ground controllers choreographing the move, Canadarm2 would painstakingly maneuver the multi-ton component – often the size of a bus – and attach it to the station.

It installed:

• The Quest Joint Airlock.
• The entire Integrated Truss Structure, segment by segment (P1, S1, P3/P4, S3/S4, P5, S5, P6, S6). This included lifting and deploying the station’s massive solar arrays.
• The Harmony (Node 2) module.
• The European Space Agency (ESA)’s Columbus laboratory.
• The Japanese Kibō (JEM) laboratory and its associated robotic arm and logistics module.
• The Tranquility (Node 3) module and the Cupola observation dome.

Without Canadarm2, the assembly of the ISS would have been impossible. It performed the heavy lifting that was far beyond the capability of spacewalking astronauts.

The Cosmic Catcher’s Mitt

After the Space Shuttle retired in 2011, Canadarm2’s job description changed from construction to logistics. It became the ISS‘s “catcher’s mitt.”

A new generation of commercial cargo vehicles was developed, most notably SpaceX’s Dragon 1 and Northrop Grumman’s Cygnus. These spacecraft, along with the Japanese H-II Transfer Vehicle (HTV), were designed to “berth” with the station, not “dock.”

Docking is a high-speed, automated process where the spacecraft flies itself directly into a port. Berthing is a more cautious, multi-stage process. The cargo vehicle flies to a “capture point” about 10 meters (33 ft) away from the station and holds its position. Then, an astronaut at the robotic workstation in the Cupola takes control of Canadarm2. They carefully “fly” the arm’s end effector to the vehicle’s grapple fixture and capture it, just as the original Canadarm captured SolarMax.

Once the arm has a firm grip, the ground controllers take over. They remotely command Canadarm2 to gently pull the multi-ton spacecraft in and guide it, with sub-centimeter precision, onto the Common Berthing Mechanism (CBM) on one of the station’s nodes (like Harmony). Once in place, bolts are driven to create a hard seal.

This “cosmic catch” is now a routine part of station operations, performed dozens of times. Canadarm2 is the essential link in the station’s supply chain, responsible for capturing the vehicles that bring new science, supplies, and food to the crew.

Dextre: The Handyman

While Canadarm2 is the station’s “arm,” it’s too large and strong for delicate work. Its “hand” (the LEE) is designed for grappling, not for holding a wrench. For fine-scale maintenance, the Canadian Space Agency (CSA) built the Special Purpose Dexterous Manipulator (SPDM), known as Dextre.

Dextre is the station’s “handyman.” It’s essentially a 3.7-meter (12 ft) tall “torso” with two smaller, highly advanced, 7-jointed arms. It has no brain of its own; it’s a “tele-robot” controlled from the ground. It usually rides on the end of Canadarm2, which carries it to a work site, or it can be “parked” on the Mobile Base System (MBS).

Each of Dextre’s “hands” is a sophisticated gripper that can also hold specialized tools. These tools are stored in a “tool belt” on Dextre’s torso, and include socket wrenches, data and power connectors, and a specialized tool for handling small objects. Dextre’s “eyes” are cameras on its arms and a main camera on its torso, giving controllers a close-up, 3D view of the work.

Dextre’s main purpose is to reduce the number of dangerous and time-consuming EVAs. It’s designed to perform “human-scale” tasks, known as Orbital Replacement Units (ORUs). These are the modular components of the space station: batteries, electrical boxes (like Remote Power Control Modules, or RPCMs), and cameras.

Since its installation in 2008 (via STS-123), Dextre’s workload has steadily increased. In 2011, it performed its first major solo task, replacing a faulty RPCM. Its most significant job has been replacing the station’s old nickel-hydrogen batteries with new lithium-ion batteries. This is a massive, multi-year project. While astronauts still perform some of the work, Dextre does much of the heavy lifting and precise alignments, swapping out the 180 kg (400 lb) batteries.

Dextre has even been used to repair Canadarm2 itself. In 2017, one of Canadarm2’s Latching End Effectors (LEEs) failed. Dextre was used, along with Canadarm2, in a complex robotic “dance” to swap the failing LEE with a spare, restoring the main arm to full function and saving the need for a complex astronaut repair.

Robots on Other Stations: Mir and Tiangong

While the Canadarm family is the most famous, other space stations have also relied on robotics.

The Soviet/Russian Contribution: Lyappa and Strela

The Soviet Mir Space Station had two types of robotic arms, though they served very different purposes.

The Lyappa arm was a unique solution to a specific problem. It was a 15-meter (49 ft) “repositioning” arm. It was used to move entire modules, like the Kristall module, from the “front” docking port to a “side” port. The arm would attach to the module, unplug it from the station, and “swing” it 90 degrees to its new location. It wasn’t a general-purpose manipulator, but a specialized, single-use construction tool.

The Strela cargo crane (“Arrow”) was a much simpler device, more like a telescoping pole. Two of them were installed on the Mir base block. They were primarily used during EVAs to help cosmonauts move equipment and themselves around the station’s exterior. They were initially hand-cranked, but later motorized. Two Strela cranes were also carried over to the International Space Station and installed on the Russian segment, where they performed a similar role before the arrival of a more advanced European arm.

The European Robotic Arm (ERA)

The third major arm on the ISS is the European Robotic Arm (ERA). Developed by the European Space Agency (ESA) and built primarily by Fokker Space (now part of Airbus Defence and Space), it had a long journey to space. It was finally launched in 2021 with the Russian Nauka module.

The ERA is an 11.3-meter (37 ft) arm. Like Canadarm2, it’s symmetrical, with “hands” at both ends, and it can “walk” between grapple fixtures. Its key distinction is that it’s designed to service the Russian segment of the ISS, which uses different grapple fixtures and operating standards than the US and International segments.

Its primary jobs are to inspect the Russian modules, support spacewalking cosmonauts by moving them around, and transfer small payloads. It’s unique in that it can be controlled from either inside the station (on a laptop) or by a cosmonaut on an EVA using a special control panel.

China’s Celestial Palace: The Tiangong Arms

China‘s Tiangong space station, a large modular station completed in 2022, is equipped with a highly capable robotic system.

The primary arm is attached to the Tianhe core module. It is a 10.2-meter (33 ft) long, 7-jointed arm that bears a strong resemblance to Canadarm2. It can “walk” around the exterior of the station on a set of grapple fixtures, and it played an essential role in the station’s assembly. During construction, it was used to grapple the Wentian and Mengtian experiment modules after they had docked, and then reposition them from the forward port to their permanent side-berthing ports.

The Wentian module carries a second, smaller arm. This arm is more precise and is used for deploying external experiments and assisting with EVAs.

The two Tiangong arms have an impressive capability: they can link together, end-to-end, to form a single, 15-meter (49 ft) long manipulator. This combined arm has the reach to inspect almost the entire station and transfer payloads from one end to the other, a capability that showcases the advancement of China‘s space robotics program.

Arms on Mars: The Robotic Geologists

While orbital arms focused on construction and servicing, a parallel branch of robotics was evolving for a different purpose: planetary science. On Mars, robotic arms became the hands and eyes of geologists, digging into the soil and placing instruments against rocks.

The Viking Landers: A First Taste of Martian Soil

The first successful robotic arms on Mars belonged to the Viking 1 and Viking 2 landers, which arrived in 1976. Each lander had a 3-meter (10 ft) articulating arm with a scoop at the end. Its job was simple but historic: dig into the red Martian regolith and deliver samples to the lander’s onboard laboratory. This lab included the Viking biological experiments, which were searching for signs of life. The arm successfully dug trenches and delivered samples, providing the first-ever in-situ analysis of Martian soil.

Mars Pathfinder and Sojourner

The 1997 Mars Pathfinder mission delivered the first rover to Mars, the microwave-sized Sojourner. Sojournerdidn’t have a full arm, but it had a small deployment mechanism for its Alpha Proton X-ray Spectrometer (APXS). The rover would drive up to a rock (like “Yogi” or “Barnacle Bill”) and press this instrument against it, giving humanity its first chemical analysis of a Martian rock.

The Robotic Field Geologists: Spirit and Opportunity

The true “robotic geologists” arrived in 2004: the Mars Exploration Rovers (MER), Spirit and Opportunity. Each rover was equipped with a 5-jointed arm called the Instrument Deployment Device (IDD).

This arm was a massive leap in capability. It was a “human-scale” arm, about 1.5 meters (5 ft) long. At the end, it held a “turret” of four sophisticated tools:

1. Microscopic Imager (MI): A “hand lens” that could take incredibly close-up, high-resolution pictures of rock texture.
2. Rock Abrasion Tool (RAT): A high-speed grinder that could brush off dust or grind a 4.5-centimeter (1.8 in) diameter hole into a rock’s surface, exposing its fresh interior.
3. Mössbauer spectrometer: An instrument to identify iron-bearing minerals.
4. Alpha Proton X-ray Spectrometer (APXS): To determine the elemental composition of the rock.

The arm allowed for a revolutionary scientific process. The rover would identify a target. The arm would first use the Microscopic Imager to get a “before” picture. Then, it would place the RAT and grind for an hour or two. Then, the arm would swing the turret back to the exact same spot, first placing the Microscopic Imager to see the fresh-cut surface, and then placing the spectrometers to analyze its composition. This “grind and analyze” technique, all enabled by the IDD, led to Opportunity‘s landmark discovery of hematite “blueberries” and sulfate-rich rocks at Meridiani Planum, which was conclusive evidence that liquid water had once existed on the surface.

The arms were incredibly durable. Opportunity‘s arm functioned for nearly 15 years, though not without problems. Its “shoulder” joint began to fail, and engineers at NASA’s Jet Propulsion Laboratory (JPL) had to develop complex workarounds, like stowing the arm in a specific way each night, to keep it alive.

Phoenix: Digging for Ice

The Phoenix (spacecraft) lander, which arrived near the Martian north pole in 2008, was a dedicated “digger.” Its Robotic Arm (RA) was a 4-jointed, 2.35-meter (7.7 ft) long manipulator. At its end was a scoop with a motorized rasp to cut into hard soil.

The arm’s mission was to dig trenches and deliver samples to two instruments on the lander’s deck: the TEGA(a miniature oven) and the MECA (a “wet chemistry” lab). The arm’s work led to the mission’s most celebrated discovery. While digging a trench, the scoop uncovered a bright white material. In images taken over several days, this material visibly vanished – it sublimated, like dry ice. It was the first definitive, visual confirmation of water ice just beneath the Martian surface.

Curiosity: The Mobile Chemistry Lab

The Mars Science Laboratory (MSL) rover, Curiosity, landed in 2012. It’s a car-sized rover with a 2.1-meter (6.9 ft) robotic arm that is an order of magnitude more complex than Spirit‘s or Opportunity‘s.

The arm’s turret alone weighs 30 kilograms (66 lbs) and carries a powerful suite of tools. It has an advanced APXS and the Mars Hand Lens Imager (MAHLI), a color microscope. But its most powerful tools are for sample collection. The Powder Acquisition Drill System (PADS) is a percussive (hammering) drill that can bore deep into Martian rock. The arm also holds CHIMRA, a scoop and a complex system of chambers and sieves to process the collected rock powder or soil.

Curiosity‘s arm performs a new, more complex task. It doesn’t just analyze the rock on the spot; it delivers the sample inside the rover’s body. After drilling, the arm retracts, rotates the turret, and carefully sieves the powder before dropping a tiny, aspirin-sized portion into inlet funnels on the rover’s deck. These funnels lead to the SAM and CheMin instruments, the most powerful chemistry labs ever sent to another planet. This is the core of Curiosity‘s mission, and it’s what allowed it to discover complex organic molecules and determine that Gale Crater was once a habitable freshwater lake.

Like Opportunity, Curiosity‘s arm has faced challenges. In 2016, the drill’s “feed” mechanism (which pushes the bit forward as it spins) became stuck. For over a year, drilling was suspended. Engineers at JPL developed an entirely new drilling technique called “Feed-Extended Drilling.” In this method, the arm itself provides the “push,” essentially “pecking” at the rock, while the drill bit spins. This innovative workaround saved the drilling component of the mission.

InSight: Listening to the Heart of Mars

The InSight lander, which arrived in 2018, used its arm for a completely different purpose. It wasn’t a science arm; it was a deployment arm.

The Instrument Deployment Arm (IDA) was a 1.8-meter (5.9 ft) arm with a simple claw-like gripper. Its mission was to perform the most delicate deployment in planetary history. It had to pick up two highly sensitive instruments from the lander’s deck and place them perfectly on the Martian surface.

1. SEIS: The seismometer, which had to be perfectly level. The arm first placed the seismometer, then carefully placed its wind and thermal shield over it.
2. HP³: A heat-flow probe, nicknamed the “mole.”

The arm’s work didn’t end there. The “mole” failed to dig itself into the ground, getting stuck in the unexpectedly clumpy soil. The InSight team improvised a new role for the robotic arm. Over many months, they used the arm’s scoop to “push” and “tamp” the soil around the mole, and even to press down directly on the probe itself, providing the friction it needed to dig. It was a creative rescue mission, controlled from 100 million miles away, that was only possible because of the arm’s versatility.

Perseverance: The Sample Cacher

The Mars 2020 mission’s Perseverance rover has the most advanced robotic arm ever built for planetary exploration. It’s a 2.1-meter (7 ft) arm with a large turret, similar to Curiosity‘s, but its purpose is far more ambitious.

The turret contains advanced science instruments, including SHERLOC (an ultraviolet spectrometer) and PIXL(an X-ray spectrometer), which map mineralogy and chemistry at a fine scale. But its primary tool is a new kind of coring drill.

Perseverance‘s mission is the first step in a Mars Sample Return campaign. The arm doesn’t just drill for powder; it drills a pristine, chalk-sized core of rock and collects it in a special sample tube. This is where a second, internal robot comes in.

Inside the rover’s “belly” is a 0.5-meter (1.6 ft) Sample Handling Arm (SHA). The main arm’s drill presents the filled sample tube to the SHA, which retrieves it. The SHA then moves the tube to an inspection camera, then to a sealing station (to hermetically seal the sample), and finally into a storage rack. It is a full-blown robotic assembly line on Mars.

The main arm also has the job of “caching” these tubes, dropping collections of them on the surface at “sample depots” for a future “fetch rover” to pick up and rocket back to Earth. This intricate, multi-robot coordination between the external arm and the internal arm is the most complex robotic system ever operated on another world.

Comparison of Key Robotic Arms

This table provides a high-level comparison of some of the most significant robotic arms discussed.

Reaching Beyond Mars: Arms on Moons, Comets, and Asteroids

Robotic arms haven’t been limited to Mars and Earth orbit. They have also been designed to perform the difficult task of “touching” small bodies like asteroids and comets, where microgravity makes landing and sampling a major challenge.

Asteroid Sample Return: Hayabusa and OSIRIS-REx

The Japan Aerospace Exploration Agency (JAXA) pioneered the “touch-and-go” (TAG) technique. Their Hayabusa (spacecraft) mission, which reached asteroid Itokawa in 2005, had a long, horn-like “sampler horn” instead of a traditional arm. The spacecraft descended, touched the horn to the surface, fired a small projectile, and “sucked up” the resulting debris. Despite numerous setbacks, it returned a few precious grains to Earth.

Hayabusa2 (2014-2020) used a similar, more advanced system to collect samples from asteroid Ryugu.

NASA’s OSIRIS-REx mission to asteroid Bennu featured a true robotic arm, the Touch-And-Go Sample Acquisition Mechanism (TAGSAM). This was a 3.35-meter (11 ft) arm with three joints. At its end was a large, round “sampler head.”

The challenge was that Bennu was a “rubble pile” asteroid, loose and fluffy. A drill or scoop wouldn’t work. The TAGSAM head was designed to act like a reverse air filter. In 2020, the OSIRIS-REx spacecraft slowly descended, and the arm extended, touching the head to Bennu‘s surface. The instant it made contact, it fired a puff of high-pressure nitrogen gas, “blasting” regolith and small pebbles into the collection head.

The maneuver was too successful. The arm punched deep into the soft surface and collected so much material (an estimated several hundred grams) that a small pebble lodged in the collector’s Mylar flap, jamming it open. The team watched in horror as precious sample material leaked out into space. They had to act fast. They canceled the planned “spin” maneuver (to weigh the sample) and commanded the TAGSAM arm to immediately move the collector head and insert it into the Sample Return Capsule (SRC). The arm successfully stowed the sample, and the SRC returned to Earth in 2023 with a historic collection of asteroid material.

China’s Lunar Ambition: Chang’e 5

In 2020, China launched the Chang’e 5 mission, a highly complex lunar sample return mission. The lander was equipped with two collection systems: a drill for subsurface samples and a robotic arm for surface samples.

The arm was a 4-jointed manipulator with a scoop at the end. In a rapid, 48-hour surface operation, the arm extended, scraped, and scooped up lunar regolith, then lifted and deposited it into the sample container on top of the ascent vehicle. This arm, along with the drill, collected 1,731 grams (3.8 lbs) of Moon rock and soil, the first lunar sample return since the 1970s.

The Future of Orbital Robotics: Maintenance and Assembly

The next generation of robotic arms is focused on autonomy, intelligence, and the growing business of “on-orbit servicing.”

On-Orbit Servicing: The New Frontier

Thousands of satellites are in orbit. When they run out of fuel or a single component breaks, they become space debris. The future of space robotics is to fix and refuel these satellites, a task that requires highly dexterous arms.

Northrop Grumman’s Mission Extension Vehicle (MEV) is an early example. It’s a simple “jet pack” that docks with a client satellite and provides propulsion. But more advanced systems are in development.

This concept is being pursued commercially by companies like MDA and GITAI, a Japanese startup that has tested robotic arms both inside and outside the ISS. The U.S. Defense Advanced Research Projects Agency (DARPA) is also funding the Robotic Servicing of Geosynchronous Satellites (RSGS) program.

The Lunar Gateway’s Canadarm3

The next great leap for the Canadarm family will be Canadarm3. This system is Canada‘s contribution to the Lunar Gateway, the small space station that will orbit the Moon as part of the Artemis program.

Canadarm3 is a complete system: a large, 8.5-meter (28 ft) arm (like Canadarm2), a smaller, more dexterous “hand” arm (like Dextre), and a set of tools. Its primary difference will be its artificial intelligence (AI).

The Lunar Gateway will be uncrewed for long periods. Canadarm3 must be able to operate autonomously, controlled from Earth with a significant time delay. It will be responsible for inspecting the station, maintaining its hardware, deploying science experiments, and capturing visiting Artemis program spacecraft. It will be a fully-fledged robotic custodian for a lunar outpost.

Summary

The history of robotic arms in space is a history of evolution, moving from simple, single-purpose tools to complex, intelligent partners. The early scoops and drills of the Luna and Viking probes proved that remote science was possible. The Canadarm on the Space Shuttle demonstrated that large-scale construction, retrieval, and repair in orbit were practical, becoming an icon of human ingenuity and international partnership.

The Mobile Servicing System on the International Space Station – Canadarm2 and Dextre – built the largest structure in space and now operates as its logistical backbone, catching cargo ships and performing maintenance, saving countless hours of risky EVA time. This model has been adopted by other nations, with the European Robotic Arm and the Tiangong station‘s arms showing that complex robotics are a standard feature of any modern space station.

On other worlds, the “robotic geologist” has become our proxy. The arms on Spirit, Opportunity, Curiosity, and Perseverance have given us the ability to not just see Mars, but to interact with it: to grind, drill, scoop, and analyze its rocks and soil, rewriting our understanding of the red planet’s past. Sample-return arms on missions like OSIRIS-REx and Chang’e 5 have become cosmic couriers, extending our reach to distant worlds and bringing pieces of them home.

Today, we stand at the threshold of a new era. The next generation of arms, driven by AI and advanced autonomy, won’t just follow commands. They will be maintainers, manufacturers, and assemblers, building the infrastructure of the future in orbit around the Earth and the Moon. They are, and will continue to be, the indispensable helping hands that allow humanity to live, work, and explore ever further from home.