Robots have fascinated and preoccupied human minds for centuries - from ancient tales of stone golems to modern science fiction. Though the word "robot" was initially penned in 1920 by Karel Čapek, inventors have endeavored to create autonomous machines since at least as far back as the 4th Century BCE.
Today, robots are widely used across a variety of industries, aiding in everything from the manufacturing of vehicles to conducting surgery. According to the International Federation of Robotics, in 2015 there were as many as 1.63 million industrial robots in operation worldwide, and that number continues to grow steadily each year.
Here's a brief history of how robotics has evolved and grown from the early imaginings of 400 BC to the global resource they are today.
What exactly is a robot?
Before we get into ins and outs of the history of robots, it is probably worth taking a little time to discuss what is actually meant by a "robot".
Let's kick things off with a few definitions...
"A reprogrammable, multifunctional manipulator designed to move material, parts, tools, or specialized devices through various programmed motions for the performance of a variety of tasks." - Robot Institute of America.
"A machine that resembles a living creature in being capable of moving independently (as by walking or rolling on wheels) and performing complex actions (such as grasping and moving objects)."
And: Such a machine built to resemble a human being or animal in appearance and behavior." - Merriam Webster Dictionary.
"Any automatically operated machine that replaces human effort, though it may not resemble human beings in appearance or perform functions in a humanlike manner. By extension, robotics is the engineering discipline dealing with the design, construction, and operation of robots." - Encyclopedia Britannica.
As we can see, these definitions all agree that robots are some form of machine, which may be in a humanoid (or animal) form, or not, and which can be programmed to carry out a specialized task or set of tasks independently of human interaction.
Most modern examples tend to be guided by external control devices, or control devices that are integrated into the robot itself. They can be fully autonomous or semi-autonomous and range from fairly simple manufacturing devices on factory floors to hypercomplex robots manufactured by companies like Honda or Boston Dynamics.
The term "robot" can be traced back to the play R.U.R. (Rossum's Universal Robots), written by Czech writer Karel Čapek. The word itself has Slavonic roots, from the word, "robota", roughly meaning “servitude,” “forced labor” or “drudgery.”
The book is about a company, called Rossum's Universal Robots, that mass produces workers using the latest biological, chemical, and physiological knowledge and techniques.
These workers lacked a "soul", were incapable of love, and had no feelings, but were capable of performing tasks that humans do not want to do.
According to historians, Capek originally wanted to call these workers labori, from the Latin for "fatigue", but thought it sounded a little too bookish. His brother, Josef, suggested that perhaps he should opt instead for roboti, or "robot" in English.
What are humanoid robots?
We have already covered what is generally considered to be a robot by today's standard. But, when most people think of robots they also immediately conjure up any one of the many humanoid robots from stage and screen.
While a common theme in science fiction books and films, the reality is only really beginning to catch up with fantasy. Moving into the future, humanoid robots could generally fill the role of professional service machines by mimicking human motion and interaction.
Those that are in operation today tend to be used primarily to automate tasks in ways that lead to cost-savings and improvements in productivity.
"The humanoid robot market is poised for significant growth. It’s projected the market for humanoid robots will be valued at $3.9 Billion in 2023, growing at a staggering 52.1% compound annual growth rate (CAGR) between 2017 and 2023.
Of all the types of humanoid robots, bipedal robots are expected to grow at the fastest compound annual growth rate (CAGR) during the forecasted period. The rapid expansion of the humanoid robots market is due primarily to the quickly improving capabilities of these robots and their viability in an ever-widening range of applications." - Robotics Industries Association.
At present, humanoid robots are being employed for uses such as inspection, maintenance, and disaster response. In this capacity, they relieve human workers of potentially difficult, not to mention dangerous, tasks. They are also used to perform repetitive tasks that require precision - robots do not get tired or need breaks, making them more efficient than humans at many such tasks.
Robots can also be used to perform routine tasks in places humans have difficulty in going, such as space. There are also hopes for them to be used in the not-too-distant future as companions for the elderly and infirm, and to act as guides or receptionists in customer-facing roles.
In these roles, humanoid robots will be able to automate many different jobs that humans are currently required to perform which may be too risky or labor-intensive. As technology is continually refined over time, the role of humanoid robots will only continue to expand.
Ancient Robots: Archytas' Pigeon, Ctesibius' Clepsydra, and More
The earliest beginnings of robotics can possibly be traced back to ancient Greece. Aristotle was one of the first great thinkers who we know considered the possibility of automated tools.
The first automaton was designed in 400 BC by Archytas of Tarentum, who is today considered the father of mathematical mechanics. Archytas' "Pigeon" was a steam-powered autonomous flying machine. Its wooden structure was based on the anatomy of the pigeon, and it contained an airtight boiler for the production of steam.
In theory, the steam's pressure would eventually exceed the resistance of the structure, allowing the robotic bird to take flight.
In 250 BC, Ctesibius created a clepsydra, or water clock, sporting a number of elaborate automatons. Though water clocks had already been in use for centuries at that point, it was during this period that Greek and Roman inventors began to update the basic designs of the clocks with features like bells, gongs, and moving figurines.
Ctesibius' design allowed for the dropping of peddles onto a loud gong, effectively making it the first alarm clock, as well as an example of early automaton design.
But it wasn't just the ancient Greeks and Romans who were experimenting with robotics. There are accounts of automatons from ancient China, as in a passage in the Daoist book Lie Zi, written in the 3rd Century BC. The passage describes a singing and dancing robot that performed for King Mu of Zhou.
According to the text, the robot was built by an inventor named Yen Shih, out of wood and leather.
11th Century to the 15th Century: Humanoid Automatons and da Vinci's Knight
One of the most important inventors during this period was Ismail al-Jazari, a Muslim engineer and mathematician who lived in Upper Mesopotamia and created a large number of mechanical devices.
Al-Jazari is credited with the creation of segmental gears and is considered by many to be the father of automatons. Many of his robotic creations were powered by water and included everything from automatic doors to a humanoid autonomous waitress who could refill drinks.
Al-Jazari's influence is particularly apparent in the later work of Leonardo da Vinci. In 1495, the famous Italian artist and painter designed an autonomous knight, which featured a series of pulleys and gears that allowed it to move its arms and jaw, as well as sit up.
The humanoid robot was informed in many ways by da Vinci's own research on human anatomy and was apparently used as entertainment at parties by da Vinci's patron, Lodovico Sforza.
16th Century to the 18th Century: Flying Robots and Musical Automatons
The creation of robots for entertainment purposes became even more popular between the 16th and 18th centuries. Though these automatons were created to entertain, it's important not to treat their designs flippantly.
Many of the technologies used in these devices paved the way for more sophisticated machines later on.
One such creation was an iron eagle, made by German mathematician Johannes Müller von Königsberg, AKA Regiomontanus. Not a great deal is known about the construction of Regiomontanus' eagle, apart from the fact that it was made of wood and iron and was constructed sometime in the 1530s.
In 1708, author John Wilkins wrote an account of the robot eagle, claiming it had flown to greet the Prussian emperor and returned to Regiomontanus. Von Königsberg is also credited as having created a robotic fly that was capable of flight.
Another key figure of this time in the creation of entertaining mechanical machines was Jacques de Vaucanson. In 1737 Vaucanson created The Flute Player - a life-sized humanoid automaton that could play up to 12 different songs on the flute.
The automaton used a series of bellows to "breathe", and had a moving mouth and tongue that could vary the airflow, allowing it to play the instrument.
Vaucanson's most memorable achievement, however, was his Digesting Duck. The duck was notable not only for being an amusing device that appeared to eat and poop, but it is also often considered the first such device to utilize rubber tubing.
The 19th Century: Chess-Playing Machines and Early Experiments With Speech
The 19th Century saw the popularity of automatons soar as touring attractions and oddities, which were used to enchant and inspire audiences across the globe. A popular type of automaton at this time was the chess-playing robot.
The most famous of these creations was The Turk, built by Wolfgang von Kempelen in the 1770s, and which toured until 1854. Though it appeared as though The Turk could play chess, the device was revealed to be a fraud - it was actually operated by a chess player concealed within its box.
Despite the elaborate ruse of The Turk and similar devices, the central conceit provided the inspiration for the true chess-playing machines which would debut in the early 20th Century.
One remarkable machine from the 19th Century, which most certainly was not a hoax, was the Euphonia - a speaking, singing robot which was operated through an early form of text-to-speech technology. Euphonia was created by Austrian mathematician and inventor, Joseph Faber.
The machine featured a humanoid, feminine face connected to a keyboard, from where the face's lips, jaw, and tongue could be controlled. A bellows and an ivory reed provided the machine's voice, and pitch and accent could be altered through a screw in the face's nose.
Euphonia was the culmination of 25 years of work for Faber and debuted to audiences in 1846. Sadly, Victorian audiences were too unsettled by the machine's blank stare and spooky, whispery voice, and the device faded into obscurity.
The Early 20th Century: El Ajedrecista, Eric, and Gakutensoku
While The Turk was revealed as a fraud, the early 20th Century saw the creation of the first true chess-playing robots. Built around 1912 by Leonardo Torres y Quevedo, El Ajedrecista (directly translated as "The Chess Player") was the first real chess-playing robot and is considered by some to be a predecessor to video games.
The device was capable of playing a particular endgame (King and Rook against King) against a human opponent and featured an electrical circuit and a system of magnets that moved the pieces. It debuted at the 1914 World's Fair in Paris, to great excitement and acclaim.
1928 saw the creation of the first British robot, named Eric. Eric was created by engineer Alan Reffell and World War I veteran Captain William Richards. Operated by two people, the robot could move its head and arms and could speak via a live radio signal.
Eric's movements were controlled by a series of gears, ropes, and pulleys, and the robot reportedly spat sparks from its mouth. As an homage to the Čapek's play Rossumovi Univerzální Roboti - where the term "robot" was first officially coined - Eric had the letters R.U.R. engraved into its chest.
The following year saw the debut of the first Japanese robot - Gakutensoku. Built around 1929 by biologist Makoto Nishimura, Gakutensoku was over seven feet (2.1 meters) tall and could change its facial expressions through the movement of gears and springs in its head.
Gakutensoku's greatest achievement, however, was its ability to write Chinese characters. Sadly, the robot went missing while on tour in Germany.
The 1940s: Asimov's Laws of Robotics and the First Artificial Neural Networks
While the 1920s saw the introduction of the term "robot", it wasn't until Isaac Asimov's 1942 short story Runaround that the term "robotics" appeared in print. In this story, Asimov laid out his famous Three Laws of Robotics - that robots must not harm humans; that they must obey orders from humans; and that they must protect themselves from threats, provided their self-preservation doesn't break either of the first two laws.
Though written as fiction, these laws provided the basis for many of the ethical questions surrounding robots and autonomous technologies, and are still referred to today.
The 1940s also saw the creation of the first artificial neural networks. In 1943, Warren McCulloch and Walter Pitts proposed the first mathematical model of a neural network using electrical circuits, to better understand how neurons operate in the brain.
Their work paved the way for the first autonomous robots that could display complex behavior, thanks to the use of artificial neural networks.
In 1948 and 1949, neurophysiologist William Grey Walter created two such robots, dubbed Elmer and Elsie, to help him understand how the brain functions. Nicknamed "tortoises", the robots could respond to and move towards the light, guiding themselves to their recharging stations when their batteries were low.
The 1950s: The Turing Test and the Unimate
Another landmark moment in the history of robotics occurred in 1950 when Alan Turing outlined his test of a machine's artificial intelligence. The Turing Test has become the benchmark of AI, in that it measures to which degree a machine's intelligence is equal to or indistinguishable from that of a human.
In its simplest form, the purpose of the test is to determine whether or not a machine can think. His work created a necessary framework for the establishment of the field of Artificial Intelligence at Dartmouth College in 1956.
The 1950s also saw the creation of the first industrial robot - the Unimate. The patent for the Unimate was filed by George Devol in 1954 and featured a robotic arm capable of transporting die-cast parts and welding them into place. The revolutionary device would soon change the face of the manufacturing industry.
The 1960s: The Industrial Robot Revolution
After Devol was granted his patent for the Unimate in 1961, the application of robots in industrial settings progressed rapidly. That same year, General Motors installed Unimate on their assembly line in Ewing, New Jersey. After the success of Unimate at General Motors, it entered full-scale production in 1966.
The 1960s saw a number of innovations and expansions on the core idea of Devol's robotic, industrial arm. In 1968, Marvin Minsky, co-founder of MIT's AI Laboratory, created a "tentacle arm" - a robotic 12-jointed arm that was powered by hydraulics and could be controlled via a joystick.
Minsky's robotic tentacle was strong enough to lift a person and could easily reach around obstacles. His research paved the way for many of the soft robotics innovations emerging today.
In 1969, Victor Scheinman created the Stanford Arm, a robotic arm that is considered to be one of the first robots to be controlled exclusively from a computer. This was a huge breakthrough, as at the time Unimate operated from a magnetic drum.
Scheinman's Arm featured six points of articulation and was built entirely in Stanford's Artificial Intelligence Lab. Though used primarily for educational purposes, the Stanford Arm marked a major breakthrough for industrial machines that could be controlled via computers.
The 1970s: The WABOT-1, Industrial Innovations, and Robots in Space
The early 1970s saw the unveiling of the world's first full-scale anthropomorphic robot - the WABOT-1. The WABOT-1 was a follow-up to 1967's WABOT and was created by Ichiro Kato at Tokyo's Waseda University.
The WABOT-1 had a vision and limb control system, allowing it to navigate itself and move freely. It could even measure distances between objects. Its "hands" featured tactile sensors, allowing it to grasp and transport objects.
It also supposedly had an estimated intelligence equal to that of an 18-month-old human and marked a massive breakthrough in humanoid robotics.
The 1970s also saw the progression of industrial robotics when, in 1973, German company KUKA released the FAMULUS - the first industrial robot with six electromechanically driven axis'. The following year, Richard Hohn developed the first industrial computer to be powered by a minicomputer - The Tomorrow Tool, or T3.
In 1978 SCARA (Selective Compliance Assembly Robotic Arm) was created. Developed by the University of Yamanashi professor Hiroshi Makino, the arm could move along four axes and became a common fixture in assembly lines in the early 1980s.
The first robots to land on Mars were Viking 1 and Viking 2, which landed on the red planet in 1976. Both robots were powered by radioisotope thermoelectric generators, which generated power from the heat given off by decaying plutonium. Though the data collected by both Vikings was ambiguous, they were the official forerunners of the Mars rovers we know today.
The 1980s: Robots in the Home, the Canadarm, and Genghis
It was in the 1980s that robots officially entered the mainstream consumer market, though mostly as simple toys. One of the most popular of these robotic toys was the Omnibot 2000 by TOMY.
The Omnibot 2000 was remote-controlled and came complete with a tray for serving drinks and snacks. Another highly sought-after robotic toy from this period was Nintendo's R.O.B or Robotic Operating Buddy. R.O.B. was marketed as a robotic second player for the Nintendo Entertainment System. It could respond to six different commands, which were communicated via light flashes from a CRT screen.
The '80s saw further developments in the field of industrial robots, with Ford adding hundreds of robots to their assembly lines worldwide. The Ford Fiesta was notable for being one of the first cars in the world whose anti-corrosion sealants were injected by robots.
Robots also continued their journeys through the cosmos in the 1980s, with the launch of the Canadarm, on the Space Shuttle Columbia in 1981.
The Canadian-made robotic arm was 50 feet (15.2 meters) long and had six points of articulation. It could be controlled by a single crew member at the control station and performed 90 successful missions during its time in service.
Often considered one of the most important robots in modern history, 1989's Genghis was a hexapod robot made by researchers at MIT. Due to its small size and inexpensive materials, Genghis is credited with shortening production time and cost for future space robot designs. It was built with 12 servo motors and 22 sensors and could traverse rocky terrain.
The 1990s: The Cyberknife, the Sojourner, and AIBO
The early 1990s saw robots enter the operation theater with the Cyberknife - a radiosurgery system that could surgically treat tumors. Developed by Stanford University neurology professor, John R. Adler, the Cyberknife was a non-invasive surgical tool which tracked and targeted tumors with narrowly-focused beams of radiation.
By 2010, the updated Cyberknife was being used in 5% of all Stanford Cancer Center's treatments.
In 1996, the Sojourner became the first rover to be sent to Mars. The small, lightweight robot was brought to Mars by the Pathfinder and successfully touched down on the planet's surface in July 1997.
During its time on Mars, Sojourner explored 2,691 square feet (250 square meters) of land and took 550 images. Because of the information gathered by Sojourner, scientists were able to determine that Mars likely had a warm, wet climate once.
Sojourner's success marked the beginning of several more NASA rover missions to Mars.
The late 1990s saw the introduction of one of the most iconic robots of the 20th Century - Sony's AIBO robotic dog. Released in 1999, AIBO was one of the first robotic pets to hit the consumer market.
AIBO could respond to voice commands and chase a pink ball that came with the purchase of the robot. Earlier this year, Sony unveiled a new, revamped AIBO for the 21st Century, which comes complete with two cameras and space-mapping capabilities.
The 21st Century: The State of Robots Today
Though we're a mere 20 years into this century, robotics have already progressed to the point where they shape a great deal of our technological landscape. Many homes now have their own Roombas - robotic vacuum cleaners that can clean your floors autonomously.
We've also seen the application of autonomous or semi-autonomous drones in everything from the military to home deliveries. Other robots are also being developed to help us around the home, including bringing Michellin Star cooking talent to a kitchen, or restaurant, near you soon.
There have been so many landmark innovations in the past few years, that they easily warrant their own article. When discussing the robotic achievements of recent years, however, it would be remiss not to mention two robots in particular - Sophia and the Boston Dynamics Dog.
Sophia made headlines a few years ago when she became the first robot to be awarded citizenship to a nation. The Android robot, created by Hanson Robotics, was granted Saudi Arabian citizenship in October 2017.
The following month, she became the first non-human to receive a United Nations title, when she was named the UN Development Programme's Innovation Champion. Sophia's AI is cloud-based, which allows for deep learning, and she can recognize and replicate a variety of human facial expressions.
Boston Dynamics has been heralded in the media as a leading developer of modern robotics.
Perhaps their most famous robot is the Boston Dynamics Dog, or BigDog, which captured worldwide attention on its unveiling in 2005. It was designed to be a robotic beast of burden for military use and featured 50 sensors over its body. It was capable of carrying weights of up to 340 lbs (150 kg) and could run at an impressive 4 mph (6.4 km/h).
More recently, Boston Dynamics revealed two more headline-grabbing robots - the MiniSpot and Atlas. MiniSpot is an autonomous robot dog that can open doors by itself, while Atlas is a sophisticated anthropomorphic robot capable of running and jumping over obstacles.
Google is also working on AI that can teach robots how to move like real animals. Using reinforced learning (RL), once downloaded into quadruped robots, Google's AIs are able to mimic how dogs move.
Other researchers are also working on novel ways to build soft robots for a variety of tasks. One team has developed soft robots with flexible spines that can, allegedly, run as fast as real cheetahs.
Other soft robots are being developed which can help prevent injuries to human coworkers; aid in disease research; get into those hard-to-reach or dangerous places; and for potential deployment in space or on other worlds.
Researchers are also working on novel robots that can actually "eat" metal to generate their own energy. This is a potentially huge development for the robots of the future and could allow them to be liberated from needing batteries and other external power sources.
What does the future have in store for robots?
If Boston Dynamics is any example to go by, robotic innovations are now emerging on a near-weekly basis. As we have seen, robotics has enjoyed a long and storied history, and it would appear that we have far more to look forward to.
But what will the future of robots look like? Would we even recognize futuristic robots if we saw one now?
Like any prediction of the future, we can only really speculate at present, but the latest developments in robotics give an insight into the general trajectory of robotics in the short to medium-term. Robotic engineers are hard at work in many countries around the world designing and developing the next generation of robots.
With an apparent drive to make some of them more human-like, improving their cognition, look and feel, the way they move and interact with the world and other people are some of the main areas of development at present. Many of the latest robots show a trend towards ever-more realistic skin and hair, and embedded sensors which allow them to interact more "naturally" in the world.
But most robots will likely not resemble humans at all. In fact, the vast majority will look reassuringly, well, robotic, for many years to come.
They will likely range in size from the microscopic (nanorobotics) to humongous, depending on their intended task. For medical interventions, researchers are already working on teeny tiny robots that can easily, and safely, be injected into patients' bodies to perform some critical medical treatments.
Other tiny robots are also being developed for search and rescue operations. Military use will likely be a major arena for robotic development over the years to come.
From support and reconnaissance roles to active combat and espionage, military robots could in the future form a major part of warfare.
Getting back to humanoid robots, robot engineers are also paying close attention to many of the subtle facial movements - micro-expressions - that often go unnoticed by most people. Some are even attempting to simulate breathing and types of non-verbal communication, in order to make robots feel more real.
The combination of artificial intelligence and more realistic "bodies" will likely be another driving force in robotics for years to come. If ever realized, we may well have to come to terms with living alongside "living" and "breathing" artificial beings in our world.
Whether or not people will accept these realistic robots is anyone's guess, but it is clear we have set on a path that may one day completely change the way we think about what it truly means to be "alive".
But let's not get ahead of ourselves. Robots, and especially humanoid robots, will need to become as self-sufficient as possible. Depending on their intended role, robots of the near to distant future will need to be able to learn things for themselves, and not just be programmed.
Even relatively simple functions like helping around the home require surprisingly complex sets of instructions to perform various tasks, from walking up a staircase to gripping different objects. This is where AI will be critical.
But whether or not such AI-robots will develop a hatred for mankind while washing your dishes or ironing your clothes is something we will have to wait and see.