This is the first question in a series that I have (that follows the human organ systems). This question deals specifically with the use of the muscular and skeletal systems in a robot. Since they work so cooperatively, I'll ask them together.

I'm imagining soft-SF answers (not technical... unless it's 100% impossible).

Say bones and muscle tissue can be fabricated/ grown using stem cells. Could they be "transplanted" into a working humanoid-robot with an artificial intelligence similar to the human brain?

If so, then the robot would need to provide, through a system similar to the circulatory system (which, for all intents and purposes, we will assume is covered/ can be emulated by tubes and wires), energy and nutrients to each cell to sustain them.

I'm not trying to say that this would become a fully sustainable species of hybrid human/ robots, but it's more of a teaching tool for the robot to understand the, in this case, limits of the human skeletal/ muscular systems.

I do plan on asking about each of the major body systems as time goes on, so if you are free for some time and want to go into greater depth about any other system and how it could/ not work and why, that's your prerogative.

Don't penalize me for offering, as my question just tackles the m/s systems, but it's pretty obvious that I'll be covering each eventually, so if you have an idea and want to roll with it across other systems, feel free.


3 Answers 3


With our current medical and technological state, it is utterly impossible. Reasons are briefly listed below:

  • Muscles need a constant glucose supply. It is not as simple to provide that in a robotic setup.
  • Muscles also need a constant supply of oxygen in the form of oxygenated hemoglobin. You can't just "pump" gaseous oxygen to them and be happy. That won't work. You will need a full circulatory system with heart, veins, arteries and all.
  • Muscles also require a very detailed and precise nervous system. It is not easy to get the muscles do something accurately with robotic electric signals. You would need the complete nervous system (or its emulation).
  • Muscles also require an immune system to fight off bad guys like necrosis pathogens etc. Without these, the muscles would quickly get rotten by the action of pathogen bacteria.
  • Muscles also require a complete setup of endocrine system (glands and stuff). They need a feedback system for growth, repair and all that. Without these, they will shrivel up and die soon.
  • Like all other cells in the body, muscles produce waste products (CO2 and water) which needs taking care of.
  • Muscles have a limit of power and the best way to use their power is by attaching them to bones. Steel is far too heavy for muscles usage and muscles would give a very bad result with that.

So no. In the absence of a complete biological environment, you cannot have muscles alone in a robot.

  • $\begingroup$ Thanks for answering. I appreciate everything you've said. $\endgroup$ Commented Oct 20, 2015 at 8:18

Path 1: Biologically Inspired

You'd have to decide which path you want to follow. Remember, biological systems evolved in a particular context, the most salient characteristic being the absence of human engineering and mining capabilities. So while saber-tooth tigers and the like might have wished for claws made of carbon steel, tungsten carbide, or of graphene-based nanoplateletes there seems to have been no easy biological evolution process to amass and structure the materials required. While some biologically-generated substances are rather amazing (see limpet teeth or spider silk), intelligence-driven engineering can take a far more directed and therefore effective approach.

So you could go for biologically inspired, but physically superior versions. Think moderately bullet-proof skin, ability to jump over tall fences, sustained running at 50mph, and so forth. The only current downside to human engineering is that due to our limited ability to manipulate structures at a molecular level at the moment, our designs are rather bad at self-repair compared to natural systems, but that will likely be corrected in the coming decades, as our nanotechnologies progress and mature.

Path 2: Biomimicry

Perhaps the human technology level has not advanced enough. After all, you have to have an energy delivery mechanism, maintenance and repair down to the molecular level, waste disposal, dealing with foreign bodies and pathogens... It's a headache. Perhaps it's easier to just go with a customized biological design. At that point, you're essentially growing body parts and you're letting the cell programming take care of things like vascularization, lymphocytes, more generally cell specialization and replacement, etc.

From a story perspective, you'd have the outwardly indistinguishable robot plotline going. Which may be a plus or a minus, depending on how it's played. On the downside, integrating biological and synthetic components might be a headache at those lower technology levels, as biological entities tend to build biofilms and deposits over whatever they identify as a foreign body.

Path 3: The Robopathogen

While you could in theory design a superbody or engineer a human-indistinguishable biomimic in a vat somewhere, it may be ... um ... easier to simply take one. Imagine an alien-like scene where the Robopathogen clamps on to the face of its human victim and drives a sharp proboscis into its victim's brain, spewing its nanomachinery inside, where it will take over all higher thought processes. Perfect for infiltration, it may even have access to the victim's stored memory systems if sufficiently advanced.

  • $\begingroup$ That's an interesting idea. Thanks for taking the time to answer! I appreciate it. $\endgroup$ Commented Oct 20, 2015 at 13:05

Assuming that all the metabolic/immune system problems cited in other answers can be solved, you'll still have a tough time with controlling those muscles since we don't currently have a good neuron/silicon interface.

The Holy grail of cybernetics since the beginning has been to get neurons and silicon to interact directly and while lots of work has been done it's not that great yet. Getting single neurons to talk to a single patch of silicon has been achieved but to control entire muscle groups will require considerable advances.

Consider the reach of our nervous system in relation to our muscles. Every muscles fiber interoperates with a neuron in a vast neural net. To achieve comparable levels of muscular control, these muscles will need to have equal access to muscle fibers. There's a spectrum across which this may be achieved starting with full neuron control at the muscular level all the way to full metal-nerve control.

Full neuron control means that the muscles will be instrumented with the normal neurons with the neuron-silicon interface controlling an entire muscle. This is the easiest to wire up because the number of integration points are relatively few.

Conversely, if the muscles are instrumented with silicon nerves then mechanisms will need to be used to distribute and maintain those individual connections between nerve and muscle fiber. I suspect that there are changes to the ways that muscle fibers and nerves interact during muscle growth. If this robot is doing any kind of non-trivial exercise, the muscles will be reorganizing. This is gonna be tricky and requires a level of nanotechnology that we don't have right now.

  • $\begingroup$ +1 for the detailed debate about electronics and circuitry. You should also have detailed on other prerequisites $\endgroup$ Commented Oct 20, 2015 at 16:26
  • $\begingroup$ Thanks for the answer. I appreciate you taking the time to explain. $\endgroup$ Commented Oct 20, 2015 at 17:28
  • $\begingroup$ @YoustayIgo You covered a lot of the things I would have written about but just wanted to emphasize an area that you didn't provide much detail on. $\endgroup$
    – Green
    Commented Oct 20, 2015 at 17:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .