(An addition to answers posted already, from Is it possible to refuel the James Webb Space Telescope?):
That depends on the JWST having its tanks/plumbing/ports accessible to a robot.
(from what I have seen, I think it is a no..)
IF it does then...yes-ish:
Northrop Grumman's MEV - Mission Extension Vehicles - have demonstrated the ability to physically grab and make a hard dock with a vehicle that was not previously designed to dock with..anything.
And with that, the MEV is able to provide station keeping, attitude control and manoeuvring if required.
Successful docking paves the way for future on-orbit and life-extension services through robotics
The Mission Extension Vehicle-1 (MEV-1), the industry’s first satellite life extension vehicle, completed its first docking to a client satellite, Intelsat IS-901 on February 25, 2020. MEV is designed to dock to geostationary satellites whose fuel is nearly depleted. Once connected to its client satellite, MEV uses its own thrusters and fuel supply to extend the satellite’s lifetime. When the customer no longer desires MEV’s service, the spacecraft will undock and move on to the next client satellite.
The Northrop Grumman-built spacecraft called MEV-2 docked successfully with the nearly 18-year-old Intelsat IS-10-02 satellite, in a move that is expected to add another five years of life to the satellite.
( Intelsat’s IS-10-02 satellite as MEV-2 approached for docking )
The shortcomings of MEV have been highlighted already - taking control of a satellite in EO is quite different from doing so at L2.
Presumably it would not be an inconsiderable technological leap to get to a workable stage at that distance, especially within 10 years.
The next step is refueling:
Robotic Satellite-Refueling Test Resumes on Space Station, 2013
Robotic Refueling Mission calls for Dextre, which sits at the end of the orbiting lab's huge Canadarm2 robotic arm, to perform simulated refueling and repair tasks on a washing-machine-size platform affixed to the station's exterior.
The experiment's goal is to demonstrate technology that could someday fix and refuel orbiting satellites robotically, thereby extending their lives and potentially saving satellite operators billions of dollars over the long haul. Such work can be challenging, since current satellites were generally not designed to be serviced.
The first RRM experiments began last year (2012), when controllers on the ground used the two-armed Dextre to snip some wires with minimal clearance. The latest round of activities will be more complex and involved, as Dextre will snip more wires, unscrew caps and pump simulated fuel, NASA officials have said.
Orbit Fab Demonstrates Satellite Refueling Technology on Space Station, 2019
Orbit Fab announced June 18 it completed tests of an experiment called Furphy on the ISS, demonstrating the ability to transfer water between two satellite testbeds. At the end of the tests, the water was transferred into the station's own water supply, the first time a private payload supplied the station with water in that manner.
The company has acknowledged that it is getting ahead of the market, since no satellite refueling systems yet exist and most satellites are not designed to be refuelable in orbit.
More on RRM:
Robotic Refueling Mission 3 Can’t Perform Cryogenic Fuel Transfer
RRM has established a firm legacy in demonstrating satellite servicing capabilities and that on-orbit servicing is technologically ready for implementation. RRM launched in July 2011 aboard the final space shuttle flight and was the last payload to be removed from the shuttle cargo bay by an astronaut. It was subsequently mounted outside onto a Express Logistics Carrier built at NASA’s Goddard Space Flight Center in Greenbelt, Maryland. RRM demonstrated and tested the tools, technologies and techniques needed to robotically refuel and repair satellites in space that were not designed to be serviced.
About a decade ago, the Space Servicing Capabilities Office was established for:
- Advancing the state of robotic servicing technology to the point where America can routinely
service satellites never designed to be serviced,
- Positioning America to be a global leader· in on-orbit repair, maintenance and satellite clisposal,
- Supporting the development of a U.S. industry for spacecraft servicing
RRM as seen here was one project, Argon was another:
The Argon system was composed of a feature recognition camera and software designed to allow a future spacecraft to robotically dock with an non-cooperative spacecraft, thereafter be able to repair, refuel or otherwise service it.
In tests, Argon sensors record images of the spacecraft target as it moves through some pre-determined motion.
The Argon Project completed integration in November 2011 and began a ground test campaign
that will culminate in an end-to-end simulation of proximity operations, approach, and capture of a non-cooperative spacecraft target in· the Fall of 2012. The Argon team will conduct a series of increasingly sophisticated demonstrations leading up to the end-to-end test. Two different models for the GOES-12 spacecraft; a geostationary satellite that is a potential candidate for a refueling mission, have been used.
Tests were conducted in late 2011 that simulated separation distances between approximately 90 meters and 1 meter, with the targets positioned statically or with relative motion simulated by an overhead crane. Current testing underway in the ssco facility at GSFC where the relative motion between Argon and the target is simulated using robotic motion platforms.
In parallel with the Argon test campaign, the ssco is conducting a development and test program to integrate robot arm technology with a maiman ring capture tool that can be used to reach out and grab on to the target spacecraft at the point of capture. This development is ongoing in the ssco facility and will come together with the Argon system in late 2012 to conduct the end~to-end non-cooperative proximity operations and capture demonstration for a potential servicing mission.
Examples of features of a cooperative spacecraft for servicing include: docking mechanism and grapple fixtures.
Examples of features of a non-cooperative spacecraft for servicing include: Marman rings, bolt holes and nozzles.
Argon's feature recognition included looking for visual fiducials (reference points), silhouettes, edges, image correlation from a database, corners, and points on the non-cooperative spacecraft.
In summary though, it is possible, and if not now then in the near future, but it depends on the will and the cost.
In that time JWST could well be superseded.
Also, I am aware of Ludo's assertion about it being very hard to do, especially at that distance and with or without haptic feedback. I knew someone who was trying to do that with robots performing human surgery across continents (USA to Australia IIRC) - very hard.
This image suggests that there is no Marman ring that a robotic spacecraft could grab hold of without damaging something.
(I have not confirmed it but it seems both Argon and MEV were looking for Marman rings to grab on to/hard dock with)
Update, Jan 2022, regarding in space refueling service provider service Orbit Fab mentioned above:
Orbit Fab, the Gas Stations in Space™ refueling service provider and Astroscale U.S. Inc., the U.S. subsidiary of Astroscale Holdings Inc. and market leader in securing long-term orbital sustainability, today announced a commercial agreement to refuel Astroscale’s Life Extension In-Orbit (LEXI™) Servicer in geostationary orbit (GEO); LEXI is the first satellite designed to be refueled.
Under the terms of this initial agreement, Orbit Fab’s GEO fuel shuttle will resupply Astroscale’s fleet of LEXI Servicers with up to 1,000 kilograms of Xenon propellant.