Our future could be full of immortal, self-repairing robots – here’s how

By Jonathan Roberts (Professor of Robotics, Queensland University of Technology)

With a generative artificial intelligence (AI) system like ChatGPT And Stable Diffusion being the talk of the town today, it might feel like we’ve taken a giant leap closer to a sci-fi reality where AI is a physical entity all around us.

Indeed, computer-based AI appears to be evolving at an unprecedented pace. But the pace of progress in robotics – which we can think of as the potential physical embodiment of AI – has been slow.

Is it possible that future AI systems will need a robotic “body” to interact with the world? If so, will nightmarish ideas like self-improvement, shapeshifting Robot T-1000 from the movie Terminator 2 come to fruition? And can a robot be made that can “live” forever?

Energy for ‘life’

Biological life forms like us need energy to operate. We get ours through a combination of food, water and oxygen. Most plants also need access to light to grow.

In the same way, immortal robots need a continuous supply of energy. Currently, electric power dominates the energy supply in the world of robotics. Most of the robots are powered by battery chemistry.

Alternative battery types have been proposed that use nuclear waste and ultra-thin diamonds at their core. Inventors, a San Francisco startup called Nano Diamond Battery, claims a possible battery life of tens of thousands of years. A very small robot would be an ideal user of such a battery.

But a more likely long-term solution to propelling robots may involve a different kind of chemistry – and even biology. In 2021, scientists from Berkeley Lab and UMAss Amherst in the US demonstrated a tiny nanobot that can get its energy from chemicals in the liquid in which they swim.

Researchers are now looking at ways to scale this idea up into larger robots that can work on solid surfaces.

Improve and imitate yourself

Of course, immortal robots may still need occasional repairs.

Ideally, the robot will repair itself whenever possible. In 2019, a Japanese research group demonstrated research robot called PR2 tighten it screw yourself using a screwdriver. It’s like a solo operation! However, such a technique will only work if a non-essential component needs to be repaired.

Another research group is exploring how soft robots can heal themselves when damaged. A group in Belgium demonstrated how the robot they developed recovered after being stabbed six times in one of its legs. It paused for a few minutes until the skin healed itself, and then walked away.

Another unusual repair concept is using other objects the robot might find in the environment to replace damaged parts.

Last year, scientists reported how Dead spiders can be used as robot grippers. This form of robotics is known as “necrobotics”. The idea is to use dead animals as ready-made mechanical devices and attach them to robots to become part of the robot.

The proof-of-concept in necrobotics involved taking a dead spider and ‘reviving’ its hydraulic legs with air, creating a very strong gripper. Preston Innovation Lab/University of Rice

Robot colony?

From all these recent developments, it’s pretty clear that in principle, one robot might live forever. But there is a very long way to go.

Most of the proposed solutions to energy, repair, and replication problems have only been demonstrated in the laboratory, under highly controlled conditions, and generally on a small scale.

The ultimate solution may be one large colony or a horde of tiny robots sharing the same brain, or minds. After all, this is exactly how many species of insects have evolved.

The concept of ant colony “mind” has been contemplated for decades. Research published in 2019 suggests that ant colonies themselves have a form of memory not contained in any of the ants.

This idea aligns very well with one day having large groups of robots that can use this trick to replace individual robots when needed, but keep the cluster “alive” indefinitely.

Ant colonies can contain ‘memories’ scattered among many individual insects. frank60/Shutterstock

Ultimately, the terrifying robot scenario outlined in countless sci-fi books and films couldn’t have suddenly developed without anyone noticing it.

Engineering highly reliable hardware is extremely difficult, especially with complex systems. Today no engineered product can last forever, or even hundreds of years. If we do find an undead robot, we will also have the opportunity to build some shelters.

Jonathan Roberts is Director of the Australian Cobotics Centre, Technical Director of Advanced Robotics for Manufacturing (ARM) Hub, and Chief Investigator at the QUT Center for Robotics. He receives funding from the Australian Research Council. He is the co-founder of the UAV Challenge – an international drone competition.

This article is republished from Conversation under Creative Commons license. Read original article.