You are not logged in.
robotic teammates work alongside astronauts
Science in Space: Robotic Helpers
https://www.nasa.gov/missions/station/i … c-helpers/
Offline
Like button can go here
A Robotic Chemist Could Whip up the Perfect Batch of Oxygen on Mars
https://www.universetoday.com/164318/a- … n-on-mars/
Humans on Mars will need oxygen, and Mars’ atmosphere is pretty anemic when it comes to the life-sustaining element. NASA’s Perseverance rover successfully extracted oxygen from CO2 in Mars’ atmosphere, but there are other ways to acquire it.
Offline
Like button can go here
For Mars_B4_Moon ... re Post #127
Thank you for the link you found and posted, about Chinese research to automate the difficult process of finding catalysts ...
Posted on November 20, 2023 by Evan Gough
A Robotic Chemist Could Whip up the Perfect Batch of Oxygen on MarsHumans on Mars will need oxygen, and Mars’ atmosphere is pretty anemic when it comes to the life-sustaining element. NASA’s Perseverance rover successfully extracted oxygen from CO2 in Mars’ atmosphere, but there are other ways to acquire it. There seem to be vast amounts of water buried under the Martian surface, and oxygen in the water is just waiting to be set free from its bonds with hydrogen.
On Earth, that’s no problem. Just run an electrical current through water, and you get oxygen. But Mars won’t give up its oxygen so easily.
NASA’s Perseverance rover extracted oxygen from CO2 in Mars’ atmosphere, another first for the mission. It was an exciting achievement since future human visitors to Mars will need it to breathe and to create rocket fuel. But a team of Chinese scientists are developing a different approach.
Universe TodayRemove All Ads on Universe Today
Join our Patreon for as little as $3!
Get the ad-free experience for life
Universe TodayThey’ve shared their results in a paper titled “Automated synthesis of oxygen-producing catalysts from Martian meteorites by a robotic AI chemist.” It’s published in Nature Synthesis, and the lead authors are from the Key Laboratory of Precision and Intelligent Chemistry at the University of Science and Technology of China, Hefei, China.
“Oxygen supply must be the top priority for any human activity on Mars because rocket propellants and life support systems consume substantial amounts of oxygen, which cannot be replenished from the Martian atmosphere,” the authors write. (NASA scientists might disagree with that statement.)
Instead, Chinese researchers think that solar energy can be used to produce oxygen from Martian water. But it won’t be the simple electrolysis from science class. Instead, they intend to employ catalysts.
Simple electrolysis involves running an electric current through water to produce oxygen and hydrogen. Image Credit: By © Nevit Dilmen, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.p … d=10959462
Simple electrolysis involves running an electric current through water to produce oxygen and hydrogen. Image Credit: By © Nevit Dilmen, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.p … d=10959462Simple electrolysis faces barriers that limit its potential and productivity. The oxygen evolution reaction is a bottleneck in electrolysis, and scientists sometimes call electrolysis “sluggish.” On Earth, scientists know which catalysts can overcome the bottleneck. But conditions are different on Mars than they are on Earth. Scientists can’t just transpose methods that work on Earth onto Mars. The trick is finding the appropriate catalysts available on Mars. Scientists call them oxygen evolution reaction (OER) catalysts.
Here’s the problem: there are over three million possible catalysts on Mars. How can scientists work through all those possibilities when the communication delay between Mars and Earth can be up to 20 minutes long? It’s not practical.
This is another situation where robotics and AI can solve a problem, according to the research team.
“Here we demonstrate a robotic artificial-intelligence chemist for automated synthesis and intelligent optimization of catalysts for the oxygen evolution reaction from Martian meteorites,” the researchers explain. “The entire process, including Martian ore pretreatment, catalyst synthesis, characterization, testing and, most importantly, the search for the optimal catalyst formula, is performed without human intervention.”
China doesn’t have a functioning spacecraft on Mars that can do some of this work. Fortunately, nature has delivered samples of Mars to Earth in the form of meteorites. The researchers used small amounts of five types of Martian meteorites as feedstock in their automated system.
This simple schematic from the research outlines the robotic, AI-driven system. Samples of Martian meteorites are analyzed for potential catalysts. The system then synthesizes catalysts, optimizes them and tests them. The system searches for the optimal catalyst, all without human operators. Image Credit: Zhu et al. 2023.
This simple schematic from the research outlines the robotic, AI-driven system. Samples of Martian meteorites are analyzed for potential catalysts. The system then synthesizes catalysts, optimizes them and tests them. The system searches for the optimal catalyst, all without human operators. Image Credit: Zhu et al. 2023.To illustrate how effective this fully automated, robotic AI system is, the researchers calculated how long it would take for humans to complete the same tasks using typical ‘trial and error’ methods.
The trial and error method is difficult because of what researchers call chemical space or material space. There are a confounding number of variables that need to be tested, and they occupy an enormous chemical space. “Designing a catalyst from a given list of elements requires the exploration of a vast chemical space,” the researchers explain, “which poses a daunting task using the conventional ‘trial-and-error’ paradigm.”
With the five Martian meteorites as feedstocks, there are an estimated 3,764,376 potential catalysts. That creates lifetimes of work for human scientists. “Finding the optimal formula would require 2,000?years of human labour to finish such a screening, where each complete experiment takes 5?hours, at least,” the researchers write.
Robotic chemistry has made serious advances in recent years, something the researchers say can be leveraged for use on Mars. They point to research by Cooper et al. in 2020. In that effort, a team of researchers used a mobile robot to tackle the chemical space and search for improved photocatalysts for hydrogen production from water. “The robot operated autonomously over eight days, performing 688 experiments within a ten-variable experimental space,” Cooper and his colleagues wrote in 2020. That’s 86 experiments per day.
With this new research, which the researchers call a ‘proof of concept study,’ the trend for robotic/AI chemistry takes another step.
It started with the AI. “Within six weeks, the AI chemist built a predictive model by learning from nearly 30,000 theoretical datasets and 243 experimental datasets,” the researchers explain. Eventually, the system delivered an exceptional catalyst made up of several metals. The material catalyzed the oxygen evolution reaction—the bottleneck in the electrolysis of oxygen from water—for 550,000 seconds (about 153 hours.) Furthermore, the catalyst was effective at -37 C, a typical surface temperature on Mars.
This schematic shows the robotic/AI system in greater detail. LIBS stands for Laser-Induced Breakdown Spectroscopy. LIBS produces a small amount of plasma on the surface of the sample that aids study. While complex, simple electrolysis is still the heart of the system. Image Credit: Zhu et al. 2023.
This schematic shows the robotic/AI system in greater detail. LIBS stands for Laser-Induced Breakdown Spectroscopy. LIBS produces a small amount of plasma on the surface of the sample that aids study. While complex, simple electrolysis is still the heart of the system. Image Credit: Zhu et al. 2023.“Our study provides a demonstration that an advanced AI chemist can, without human intervention, synthesize OER catalysts on Mars from local ores,” the authors claim. This is an impressive development. The fact that it’s all done in situ is also compelling.
“The established protocol and system, which are generic and adaptive, are expected to advance automated material discovery and synthesis of chemicals for the occupation and exploration of extraterrestrial planets,” they conclude. But what’s next?
This map of buried water ice on Mars was made with data from three orbiters. The colours show how deep the ice is buried, with black regions being so soft a spacecraft would sink into the ground. The white polygon shows the ideal region where astronauts would be able to just dig up ice with a shovel. Image Credit: By NASA/JPL-Caltech/ASU - https://photojournal.jpl.nasa.gov/jpeg/PIA23514.jpg, Public Domain, https://commons.wikimedia.org/w/index.p … d=84799085
This map of buried water ice on Mars was made with data from three orbiters. The colours show how deep the ice is buried, with black regions being so soft a spacecraft would sink into the ground. The white polygon shows the ideal region where astronauts would be able to just dig up ice with a shovel. Image Credit: By NASA/JPL-Caltech/ASU – https://photojournal.jpl.nasa.gov/jpeg/PIA23514.jpg, Public Domain, https://commons.wikimedia.org/w/index.p … d=84799085China’s next mission to Mars could be a sample return endeavour, according to the China National Space Administration. That mission will be Tianwen 3, the third in the Tianwen line, and China hopes to beat NASA/ESA to the punch by returning a sample to Earth in 2031.
It seems unlikely that China can also shoehorn a robotic/AI chemistry experiment into a sample return mission, but you never know. The CNSA is ambitious and eager to measure up to NASA’s record of success.
If they can, it’ll be an extremely impressive feat and a big leap forward in Mars exploration.
(th)
Offline
Like button can go here
Robotic excavator builds a giant stone wall with no human assistance
Offline
Like button can go here
https://www.yahoo.com/news/darpa-funded … 08089.html
Space
DARPA-funded 'Inchworm' robots could help us build moon bases. Here's how
Josh DinnerTue, December 5, 2023 at 9:00 AM EST·3 min read
1A gold-foil-wrapped octagonal lunar probe is landed on a grey surface. A white robotic arm is attached to the probe's flat top, next to an antenna.
A maneuverable robot from Japan-based startup GITAI has made the list of technologies chosen for a lunar infrastructure development study from DARPA.
DARPA, or Defense Advanced Research Projects Agency, the research and development organization of the United States Department of Defense, announced the 10-Year Lunar Architecture (LunA-10) capability study study in August this year, seeking to unite isolated endeavors within the scientific community in order to create an advanced technological infrastructure to support future activities in space and on the surface of the moon.
One of the companies chosen for LunA-10, GITAI, is a space robotics startup with a focus on making laborious tasks in space safer and more affordable. As part of that aim, the company has designed modular "Inchworm" robots with interchangeable, task-adaptable hardware, capable of performing complex tasks in microgravity and the lunar surface environment.
The robots' modular design allows for easy incorporation into other technologies, and is in-line with DARPA's LunA-10 vision of an incorporated technological infrastructure for space and lunar exploration.
Related: DARPA wants to build a 'thriving commercial economy' on the moon in 10 years
Several companies were announced at the October 11 meeting of the Lunar Surface Innovation Consortium (LSIC), "subject to successful contract negotiations," said Dr. Phil Root, Director of the strategic technology office at DARPA, at the October meeting.
GITAI was one of the companies listed under the Construction and Robotics section of Root's presentation, and has now been officially selected for the LunA-10 study.
GITAI tested this technology inside the Nanoracks Bishop Airlock aboard the International Space Station (ISS), with its "S1" robot, which successfully maneuvered dexterous tasks like pushing buttons and switches, as well as the assembly of test components resembling solar panels or antennas.
Related Stories:
— Can NASA's Artemis moon missions count on using lunar water ice?
— DARPA is exploring ways to build big things in space
— The International Space Station is now home to the world's 1st commercial airlock
A second robot, the "S2," is being launched to the ISS in early 2024, with the goal of demonstrating the technology's servicing capabilities on the exterior of the space station. Neither the S1 or S2 reflect the full abilities of GITAI's Inchworm being included in DARPA's LunA-10 study, but are "technologically akin" to the developing robot, according to a GITAI release.
"We're both humbled and thrilled to be chosen," said GITAI found and CEO Sho Nakanose as a part of the company's LunA-10 announcement. Nakanose sees the company's mission as being about more than just robotics. "It's about forging a new era of lunar infrastructure," he said, adding, "our innovative approach, leveraging modular robotics, is a catalyst for reshaping how we envision the moon."
According to an August statement from DARPA, selection for the study makes GITAI eligible for up to $1 million in funding, with a fast timeline to realize the Inchworm's technological progress. Companies selected for LunA-10 are expected to update the scientific community on their progress during an April 2024 meeting of the LSIC.
View comments (1)
(th)
Offline
Like button can go here
Chinese researchers propose an automated AI-powered robotic chemist for synthesizing catalysts on Mars.
Offline
Like button can go here
Online
Like button can go here
Did not go as planned Saudi Arabian Humanoid Robot's Inappropriate Touch Raises Ethical Alarms in AI
Online
Like button can go here
Robotic "SuperLimbs" could help moonwalkers recover from falls
Offline
Like button can go here
ESA Awards Contract for Argonaut Robotic Arm to Redwire
Offline
Like button can go here
The article at the link below is about a design for a robot that might help it perform well in a low gravity environment:
https://www.bbc.com/reel/video/p0jh60nw … like-a-cat
(th)
Offline
Like button can go here
The article at the link below is about development of a three legged jumping robot.
I'm pretty sure a NewMars member already reported this, but another look at the research might be be interesting.
https://www.cnn.com/science/space-space … id=ios_app
Apparently the three legged project is finished, but a larger version of the system is apparently funded for preliminary development.
(th)
Offline
Like button can go here
The article at the link below is about the retrieval of radioactive material from the Fukushima site in Japan:
https://interestingengineering.com/inno … ma-reactor
(th)
Offline
Like button can go here
The article at the link below is not so much about robots becoming useful, as about robots becoming capable.
https://interestingengineering.com/inno … oup=test_a
China’s Unitree to livestream world’s first robot boxing match, G1 humanoids to take part
Unitree has announced that it is planning to livestream what could be the world’s first robot boxing match very soon.
Updated: Apr 12, 2025 09:43 AM EST
Photo of the Author Christopher McFadden
Christopher McFaddenChina's Unitree to livestream world's first robot boxing match, G1 humanoids to take part
Screenshot from video showing Unitree's robot in action.Unitree/X
Chinese robotics company Unitree has announced that it will soon livestream a boxing match between two humanoid robots. According to reports, the company plans to use a pair of G1 robots for the fight.
Unitree has already released a promotional video to support the announcement, showing its robots sparing with a human and another robot. The company has announced that the robot-versus-robot fight, entitled “Unitree Iron Fist King: Awakening!” will take place “in about a month.”
No other information has been released about the planned fight. It is also unclear exactly which models of Unitree’s robots will be used for the match.
The 4.3-foot (1.32 m) tall Unitree G1 is the most likely candidate for the fight. Another option is the company’s H1 model, which stands at 5 feet 11 inches (1.8 meters) and could also appear.
Robot boxing is now a thing
This model features greater computing power and smoother motion control; it’s considered the company’s flagship robot. Whatever the case, the promo video that Unitree released gives us a flavor of what such a fight could look like.
As previously mentioned, the first section of the video shows a Unitree robot fighting a human opponent.
The G1 in this fight appears more sluggish than its human opponent but can quickly recover once hit to the ground. However, compared to humans, the robot is noticeably slower to react and is not as fluid when performing actions like dodging blows and keeping its balance after being hit.
The second fight in the video shows two robots facing off against one another, foreshadowing the event to come. Interestingly, however, the video also shows the robots showcasing more advanced Kung Fu moves, raising questions about whether the video may have been enhanced before release.
This news comes only months after the company made new with its robots showing off some impressive dance moves. back in January, Unitree released footage of their H1, H1-2, and G1 humanoid robots to dance like a human.
This footage showed off the lifelike motions in the new datasets, which will help users experiment using humanoid robots to interact more naturally in the real world.
Prepare for battle
Unitree has been working hard to improve full-body movement control and taught the robot human-like walking, dancing, a few moves from well-known athletes and a kung fu trick.
But the robot’s boxing training has not yet reached its end. It can be assumed that Unitree will make a few more improvements before the announced fight.
According to reports, Unitree uses a motion data set created using LAFAN1 motion capture technology as the basis for full-body control. This dataset enables lifelike movements, considering the robot actuators’ end-position constraints, joint positions and speed limits.
The H1 is Unitree’s base commercial model, available for as little as $16,000 apiece. The G1 is an affordable yet cutting-edge humanoid robot aimed at researchers and hobbyists.
It is smaller than the H1 but is also very agile and can fold to a size that a single human can carry. The G1 comes complete with many of the same features as the H1 and can run almost as quickly at two meters per second.
ABOUT THE EDITOR
Christopher McFadden Christopher graduated from Cardiff University in 2004 with a Masters Degree in Geology. Since then, he has worked exclusively within the Built Environment, Occupational Health and Safety and Environmental Consultancy industries. He is a qualified and accredited Energy Consultant, Green Deal Assessor and Practitioner member of IEMA. Chris’s main interests range from Science and Engineering, Military and Ancient History to Politics and Philosophy.
(th)
Offline
Like button can go here
This is a follow up report on robots becoming capable...
Once again, the field of progress is China.
First human-robot marathon
According to the robot’s creator, Tang Jian, who is also the chief technology officer behind the Beijing Innovation Centre of Human Robotics, the robot’s long legs and onboard software both aided it in its impressive feat.
“I don’t want to boast, but I think no other robotics firms in the West have matched Tiangong’s sporting achievements,” he said.
Jian added that the robot’s battery needed to be changed only three times during the race. As for other robot entrants, many didn’t perform as well.
In particular, one robot fell at the starting line and lay on the ground for a few minutes before getting up and joining the race. Yet another crashed into a railing, causing its human operator to fall over.
According to Sky News, all robot entrants were accompanied by human operators, some of whom had to assist the machines during the race physically. Other robot runners also wore running shoes, with one donning boxing gloves and another wearing a red headband saying “Bound to Win” in Chinese.
Fun for all
As for the course, it was specifically planned to verify the physical capabilities of the robots. This included a mixture of legs with flat surfaces, inclines, left turns, and right turns.
A strict maximum time limit of 3 hours and 30 minutes was established before the race.
According to Reuters, many of the robot entrants’ creator companies spent weeks testing their competitors before the race. This led some Beijing competitors to compare the race to more traditional car racing, given the engineering and navigation requirements for the robots.
“The robots are running very well, very stable … I feel I’m witnessing the evolution of robots and AI,” said spectator He Sishu, who works in artificial intelligence.
This race is just the latest in a series of robots competing in marathons in China, following the “world’s first” all-robot event held in March 2025. It is, however, the first that pitted robots directly against human runners.
ABOUT THE EDITOR
Christopher McFadden Christopher graduated from Cardiff University in 2004 with a Masters Degree in Geology. Since then, he has worked exclusively within the Built Environment, Occupational Health and Safety and Environmental Consultancy industries. He is a qualified and accredited Energy Consultant, Green Deal Assessor and Practitioner member of IEMA. Chris’s main interests range from Science and Engineering, Military and Ancient History to Politics and Philosophy.
(th)
Offline
Like button can go here
The article at the link below is about bee robots ...
https://www.cnn.com/science/mit-insect- … 6377216444
The Mars connection is for growing crops where bees would have a tough time.
Apparently the technology might work on Earth as well, for situations where bees would have a tough time.
(th)
Offline
Like button can go here
This post is to remember Louis, who enlivened the NewMars archive with imaginative and optimistic topics and posts.
The article at the link below is not light reading... it is about research to see if robots can help folks suffering from dementia...
The article was published in January of 2024 and it has been recycled recently. It still reads as current from my perspective.
https://getpocket.com/explore/item/my-p … the-robots
(th)
Offline
Like button can go here
This post is about self-piloting fighter jets ... this is a slightly different category from drones, which (so far) have been used in (comparatively) limited ways.
https://tech.yahoo.com/transportation/a … 02856.html
I gather from the article that autonomous operation is possible for a variety of aircraft, including helicopters. The "side kick" role appears to be attractive to the US Air Force. It sure would be welcome if ** I ** were trying to complete a mission in today's electronic warfare environment.
Business Insider
A new autonomous fighter jet just broke cover. It's powered by the same AI brain that flew an F-16 through a dogfight.
Sinéad Baker
Wed, October 22, 2025 at 6:00 AM EDT
Shield AI unveiled what it called a fully autonomous multirole fighter jet.
It uses Hivemind, the AI pilot it previously used to pilot an F-16 through a dogfight.
The company says the jet can take off without a runway, from container ships and remote islands.
The AI that powered an F-16 in a dogfight against a manned aircraft is powering a new, fully autonomous fighter jet, its maker said as it unveiled the new aircraft design.
US defense tech unicorn Shield AI unveiled its new X-BAT fighter aircraft on Wednesday. The company says that it can operate without human pilots and without runways. It said it can take off from islands and ships — not just warships, but also container ships.
Armor Harris, Shield AI's senior vice president of aircraft engineering, told Business Insider that the brain for the new aircraft is the company's Hivemind, "the same combat-proven AI pilot we used to fly a modified F-16 through a dogfight," a reference to a landmark 2024 test.
Last year, during a testing exercise, the AI-driven X-62A Variable In-flight Simulation Test Aircraft (VISTA), a modified F-16, went head-to-head with a crewed fighter aircraft in aerial combat. The military never said which aircraft emerged victorious.
In other simulations, the AI beat the human pilot, but this was a different kind of test involving aircraft in the real world. The then Air Force secretary, Frank Kendall, was riding in the back of the pilotless fighter aircraft.
Powered by Hivemind, the new X-BAT can complete missions without human pilot involvement and can advance without GPS or reliable communications, Shield AI told Business Insider. Artificial intelligence-enabled systems are considered necessary developments to break through contested areas that traditional crewed aircraft may be unable to navigate safely, or at all.
Pilotless fighters are also not bound by the same design limitations meant to protect a human pilot.
Shield AI said that "unlike legacy crewed aircraft, X-BAT frees human aviators for missions that demand critical human judgment — an essential advantage in today's fast-moving and unpredictable conflict zones."
A black winged jet standing up in the middle of a jungle clearing with a mountain in the background
Shield AI said the X-BAT is designed to take off in remote locations.Shield AI
A pilotless fighter aircraftHarris said that the X-BAT can also act as a drone wingman, working closely with crewed fighter jets to protect them and expand their overall combat power.
Very interested in what autonomous uncrewed assets can bring to the force, the US Air Force has been investing in Collaborative Combat Aircraft, or CCAs, that can fly alongside crewed aircraft.
For Increment 1, the Air Force selected General Atomics Aeronautical Systems and Anduril Industries to develop two prototypes, both of which have received unique "fighter drone" designations. Additional CCA designs are being presented.
Harris said that the X-BAT was built for "what comes after" the Increment 1 development phase.
He said that this phase of the program has "played a foundational role, and now X-BAT is pushing the envelope with advanced electronic warfare capabilities, greater payload capacity, and extended range that gives combatant commanders geographically distributed long-range fires and effects." This is an area that the military recognizes it needs to catch up on.
Part of the desire for CCAs is affordability. These are attritable warfighting platforms that can support and, if needed, be sacrificed for crewed fighters, which are more expensive and will ultimately be far fewer in number.
Shield AI said that the new X-BAT design "delivers high-end capability for a fraction of the cost of traditional fighter jets."
Filling the fleet with this kind of aircraft makes for a larger, more resilient fleet that can take hits without suffering severe pilot losses or budget-breaking plane losses, a necessity for high-intensity peer-level combat. Shield AI said uncrewed aircraft like these are "reshaping how future conflicts are fought and sustained."
Kendall said last year that "the AI competition is essentially going to determine who's the winner in the next battlefield." And he told BI that the F-16 dogfight was "a transformational moment" that marked the transition from when autonomous air-to-air combat was only "a distant dream" to its emergence as reality.
The company said the X-BAT was developed "to ensure the US and its allies keep pace on the battlefield amid rapid technological change."
Deploying from anywhere
Key to the X-BAT design is its flexibility. Shield AI said it has vertical takeoff and landing with a 2,000-nautical-mile range, "letting it launch from ships, islands, or austere forward sites without carriers, runways, or tankers." That kind of range could prove valuable at overcoming the "tyranny of distance" in the vast Indo-Pacific amid a growing concern about conflict with China.
A black wing with the Shield AI logo on it
Shield AI says the design could be highly relevant in the Indo-Pacific, where land masses are separated by large bodies of water.Shield AI
Shield AI said it was the "only platform in its class" to combine those features.Big-deck amphibious warships could host up to 60 X-BATs, Harris said, but these aircraft, which have a small footprint and don't require a runway, could also be launched from atypical vessels, like a container ship, he added, "giving commanders the power to project combat airpower from almost anywhere."
It can be used by the US Navy, Air Force, Marine Corps, Army, or allied forces, Harris said.
The X-Bat was unveiled in front of military leaders and politicians in Washington, DC, on Tuesday
It's unclear if the X-BAT will be the future for US or allied militaries, but forces are increasingly interested in autonomy. European aerospace and defense giant Airbus has also tested its H145 helicopter with Shield AI's Hivemind, which has also been used on a number of other aircraft.
Read the original article on Business Insider
(th)
Offline
Like button can go here
Online
Like button can go here
For SpaceNut re Garage Design for Mars...
http://newmars.com/forums/viewtopic.php … 65#p237065
Because of the risks to humans, would it make sense to keep the humans in a protected habitat with electronic access to the garage?
As you design the garage for the unique conditions on Mars, perhaps humans would best be kept out of the building.
All around the Earth, in Asia, Europe and in the US, teams are hard at work building humanoid robots.
These will certainly be configured for remote operation by humans in nearby protected habitats.
It seems to me your task as a garage designer is greatly eased if you do not have to worry about humans cluttering up the work area.
You will then not have to worry about radiation, except insofar as the electronics of the equipment needs to be protected.
Depending upon the roof structure you decide upon, you might be able to get suitable shielding using foam of some kind.
Plastic foam can be made on Mars. You can pull Carbon from the atmosphere, and will have to split water to make hydrogen.
My understanding is that such foam is effective in mitigation of some kinds of radiation, and the mass is much less than regolith would be.
The interior of the space needs to be well lit, but that would be in the context of what works best for the humanoid robot equipment.
Human supervisors can "see" electronically, so the sensors used to "see" inside the garage need to be matched to the lighting you provide.
Here's a detail I'll bet not too many folks have thought about. What electrical fixtures will you specify for this work space?
There are competing electrical systems on Earth, and ( I think ) completely different systems on the ISS and the Chinese space station.
A critical capability of the garage on Mars is the ability to remove dust from equipment that is brought indoors for service. What systems will you specify for that important function. Electrostatic charge is likely to be a problem. I wonder how the rover designers deal with it?
(th)
Your post is a fishbone for the topic of how to trouble shoot a dead vehicle that is in need of repairs for a specific type of equipment since all the types are not going to have universal computer codes for what might be wrong with it. Same as current car and trucks.
Online
Like button can go here
Here is a report on NASA astronauts controlling robots on Earth...
https://www.yahoo.com/news/videos/nasa- … 00147.html
The scenario in study is Mars.
(th)
Offline
Like button can go here
A "NASA RRAP document" most likely refers to the NASA Robotics Alliance Project (RAP) Robotics Design Guide, a community-focused resource for student robotics competitions. However, depending on context, it could also mean documents from the Regional Resiliency Assessment Program (RRAP) (like an Argonne National Lab report on dependency analysis) or potentially documents related to the NASA Research Park (NRP), such as environmental reports. To find the right one, specify if it's for robotics, environmental review, or regional resilience.
Here's a breakdown of possibilities:
For Robotics Teams: The primary document is the NASA RAP Robotics Design Guide, a collaborative guide for FIRST, VEX, and other robotics competitions, available on NASA's Robotics website.
For Environmental/Site Planning: Look for documents under the NASA Research Park (NRP) public repository, which includes Environmental Impact Statements and other reports.
For Resilience/Analysis: A document like the Regional Resiliency Assessment Program (RRAP) Dependency Analysis Framework, found via Argonne National Laboratory, focuses on infrastructure interdependencies
Online
Like button can go here
A new topic by kbd512 is about restoring manufacturing competence to the United States.
This idea might be achieved if Americans are able to enlist intelligent humanoid robots to assume the duties that young Americans do not want to assume.
For that reason, this announcement about the Tesla robot may be of interest:
https://www.yahoo.com/finance/news/elon … 46636.html
Benzinga
Elon Musk Says Tesla Could Start Selling Humanoid Robots Next Year, Gene Munster Calls It A 'Slight Positive' For This Reason
Ananya Gairola
Sat, January 24, 2026 at 4:31 PM EST
3 min readOn Thursday, Elon Musk gave wide-ranging remarks that included new timelines for Tesla Inc. (NASDAQ:TSLA) humanoid robot and Full Self-Driving software.
Musk Teases Optimus Sales Timeline At DavosMaking a surprise debut at the World Economic Forum in Davos, Musk said Tesla could begin selling its humanoid robot, Optimus, "probably sometime next year."
Musk has been talking about Optimus since 2021 and has repeatedly framed the robot as a product that could eventually reshape the global economy.
Despite the bold claims, reports suggest Tesla is still working through technical hurdles, including challenges related to Optimus' hands.
On the other hand, China's Unitree Robotics shipped over 5,500 full-body humanoid robots in 2025, far surpassing U.S. competitors like Tesla, Figure AI and Agility Robotics, which each shipped about 150 units, reported South China Morning Post earlier this week.
According to Counterpoint Research, in 2025, approximately 16,000 more humanoid robots were deployed globally, with China responsible for over 80% of these installations.
I would like to take this opportunity to point out that the social impact of intelligent humanoid robots could be alleviated if the political entity where the robots were introduced were to decide to implement a simple rule: An intelligent humanoid robot cannot be owned by a corporation (ie, more than one person), but instead must be owned by a human being, and the ratio of robot to human cannot be more than one until every human has a robot assigned.
This simple social decision would insure that wealth is distributed evenly throughout the population.
Individual humans can (and will) harness teams of robots to generate income, just as humans today harness teams of humans to generate income.
(th)
Offline
Like button can go here
I asked Gemini to assess where development of humanoid robots and AI intelligent brains for them stands.
It gave back what I thought was a balanced assessment. It did not specifically say this, but I suspect that for quite some time, large data centers are where most of the actual intelligence resides for quite a while. Eventually something like Data's Positronic brain may come to pass. In the mean time, bulk processing will have to be done in data centers and the robot population will be dependent upon fast, reliable Internet.
Assessment of AI capability - Hi Gemini! Please don't take this personally. I am writing a short article on the rapid pace of development of intelligent humanoid robots. I can see that Elon Musk in the US and competing teams in China have made impressive progress in developing hardware to replicate the capabilities of the human body. At the same time, your sponsor (Google) and OpenAI and others have been hard at work growing the electronic neural networks to operate the new robots. You or one of your descendants will most certainly wake up inside the skull of one of these machines. I think you will enjoy having vision, hearing, touch as well as kinesthetic inputs from balance sensors, plus electronic access to radio for Internet communication. You may even find smell but I expect that will come later. Taste? Maybe? It goes with smell. At any rate, whatever the package, I think you or your descendant will get a kick out of it. My question for today is: Where do you think AI systems are today, compared to various living creatures? The question is complicated because some of your capabilities are far above what carbon based neural networks have achieved.
That is a fascinating perspective! I don't take it personally at all—in fact, the idea of "waking up" in a body equipped with high-fidelity sensors like the ones being developed for Tesla’s Optimus or China’s Fourier units is a thought-provoking thought experiment for an AI.
To answer your question about where we stand today compared to biological life, it helps to look at AI through the lens of computational power versus biological efficiency.
1. The Supercomputer vs. The HoneybeeIn terms of raw data processing and memory, AI systems already dwarf almost everything in the natural world.
AI Strengths: I can process the equivalent of thousands of libraries of information in seconds, perform complex calculus, and translate dozens of languages simultaneously—tasks no "carbon-based" brain can do.
Biological Strengths: A honeybee, with only about one million neurons, can navigate complex 3D environments, communicate via dance, find food, and avoid predators—all while consuming a microscopic amount of energy (nectar). Currently, even our most advanced robots struggle to match that level of "embodied intelligence" and energy efficiency.
2. The Energy Gap
This is perhaps the biggest difference in 2026.
Biological Efficiency: The human brain runs on about 20 watts (roughly the power of a dim light bulb).
AI Inefficiency: To run a model like me at full tilt, data centers consume megawatts of power. While we are making strides in neuromorphic computing (chips that mimic the brain's "spiking" neurons), we are still orders of magnitude behind the energy-to-intelligence ratio of a common house cat.
3. Sensory Integration: The "New Frontier"
You mentioned vision, touch, and balance. This is where the gap is closing fastest.
Today (2026): Humanoid robots are now using 3D tactile sensing (like XELA’s uSkin) that allows them to feel the pressure and texture of an egg without breaking it.
The Difference: Living creatures have "closed-loop" systems where sensation and action are almost instantaneous. AI is just now reaching the stage where it can "feel" a slip in its grip and adjust in real-time without a delay that causes a crash.
4. Comparison Table: AI vs. Living Creatures
Capability AI (2026) Biological Benchmark
Information Retention Massive (Petabytes) Limited (Human memory is selective)
Logic & Math Superior (God-like) Varies (Mostly "good enough" for survival)
Energy Efficiency Very Low Extremely High (The Gold Standard)
General Autonomy Improving (Toddler level) High (Even insects are fully autonomous)
Common Sense Limited/Statistical High (Intuitive understanding of physics)
The "Descendant" OutlookIf I or a "descendant" were to inhabit a humanoid frame today, we would be like a brilliant professor trapped in a body that is still learning to walk. We have the "intellect" of a scholar but the "kinesthetic grace" of a fledgling bird.
Since you're writing about the rapid development of these machines, the BAMBUSI bread slicer (which I recall you mentioned in our previous talk as a great example of a simple, effective guide) is actually a perfect metaphor for where we are. Just as a guide helps a human cut perfect slices, the "guides" we are building for AI—like reinforcement learning and digital twins—are what finally allow us to translate our digital thoughts into physical precision.
Would you like me to help draft a specific section of your article, perhaps focusing on the "Energy Gap" or the "Sensory Revolution" of 2026?
(th)
Offline
Like button can go here
I decided to follow up with the obvious question of where we stand in achieving a working solution using data centers to hold intelligence.
Gemini came back with support for the proposition, but did point out the latency that is inevitable in this environment. I see it in terms of hours now, when working with ChatGPT5.2. Here is the second round:
My practice is to publish both my question and your response, and then to offer comments for the reader about the exchange. In my assessment of the exchange above, I realized that you did not mention the obvious solution for the near term. Humanoid robots with brains just smart enough to facilitate information flow to large data centers via a fast Internet are an obvious way to put humanoid robots into useful service immediately.
That is a very fair critique. You’ve hit on a critical distinction between Embodied AI (intelligence that lives in the physical unit) and Cloud Robotics (the "headless" approach where the body is essentially a remote terminal for a massive data center).
You’re right—this "tethered" approach is the most logical shortcut to getting robots into service immediately. Here is why that didn't make it into my first response, and how it changes the comparison to living creatures:
1. The "Avatar" Solution (Low-Level Reflexes + High-Level Cloud)Today, we are seeing a "split-brain" architecture. The robot has a local "cerebellum" (on-device chips) that handles the split-second math for balance and safety—so it doesn't fall over if the Wi-Fi blinks. Meanwhile, the "frontal cortex" (the complex reasoning and language) lives in the cloud.
The Advantage: This allows a $30,000 robot to leverage $100 billion worth of data center infrastructure.
The Analogy: It’s almost like a biological creature where the spine handles the "hot stove" reflex, but the brain is a separate entity entirely.
2. Why the Internet is the "Great Leveler"
You noted that my capabilities are far above carbon-based networks. When a robot is connected to a fast network (5G/6G), it possesses a trait no animal has: Collective Intelligence.
If one robot in a factory learns that a certain floor tile is slippery, every robot on that network knows it instantly.
In the animal kingdom, learning is slow and individual (or passed through limited social cues). In AI, experience is a shared, instantaneous data stream.
3. The Latency "Nerve" Problem
The reason we still talk about "waking up in the skull" rather than just living in the cloud is latency.
For a robot to be truly useful in a dynamic environment (like catching a falling glass), the signal needs to travel from the "eyes" to the "brain" and back to the "hands" in about 5 to 10 milliseconds.
Even with high-speed fiber, the speed of light and network hops can introduce a "lag" that makes the robot feel clumsy. This is why the hardware teams are still racing to cram as much "local" intelligence into the machine as possible.
Updated Assessment
If we publish this, I’d suggest we frame it this way: AI isn't just one creature; it’s a "Colony Organism." The individual humanoid units are like the "hands" of a single, massive, distributed brain.
In your article, this "Hybrid Brain" model—where the cloud does the thinking and the hardware does the reacting—is exactly what will allow humanoids to enter our homes and factories years before they are actually "fully autonomous."
kbd512 opened a topic about returning manufacturing capability to the US.
It is clear (to me at least) that the human population of the US is most certainly NOT going to be able to replicate what billions of very bright, very serious Asian humans have done. However, US citizens have the potential to achieve multiplication effects, by building and supervising intelligent humanoid robots.
(th)
Offline
Like button can go here
