When you think about the term robotics, you start thinking of some kind of human-like robot from science fiction, probably made of metal or some futuristic polymer and with possibly good, possibly bad intentions towards humankind.
Or perhaps you’re imagining a sterile production line of precision robot arms busily working on a car assembly line in some high-tech factory or grabbing Amazon items from endless shelves in a huge warehouse somewhere.
It might come as a surprise then to find out that current advances in robotics are now gathering pace to such an extent that science fiction is beginning to become science fact. Robotics is rapidly moving out of the factory floor and into the real world. In fact, it’s likely that we’re just a matter of years away from an increasing presence of robotic devices and automated semi-intelligent machines in our workplaces, on our streets, and even in our homes.
The age of robotics has arrived and here’s what you need to know.
Robotics: The Basics
While the definition of robot may vary, it’s generally agreed upon that a robot is a machine capable of carrying out tasks on its own (i.e. without a human operator) based upon a set of instructions that’s been provided by a computer. Sometimes a human is still involved, either in operating a robot or providing oversight to what it does, but even in these instances the same base-level programming applies in one form or another with full-autonomous action by the machine becoming continually more common.
The term robot originates from the Czech word ‘robota’ which literally translates as something similar to “forced labour”. The actual word was used for the first time in the 1920 science fiction play R.U.R (Rossum’s Universal Roberts) by Czech writer Karel Čapek in which he references artificial humanoids.
Until now (and at least for the time being), this definition of robot has served as a pretty fair description for the pre-programmed machines that we use to automate jobs that are either repetitive, dangerous, expensive (for human labor), or a combination of all three. As technology and developments in the fields of machine intelligence, data processing, and materials science continue to rapidly advance however, this literal definition of robot may need to be revised as the machines we create increasingly carry out tasks independently of humans.
The term robotics refers to the branch of technology, research, and development that deals with the designing, building, and operating of robots. As a field, robotics encompasses multiple disciplines from mechanical and electrical engineering through to data and computer science. Practically, robotics is the physical extension of computer controlled, programmed machines, or artificial entities that operate with minimal human involvement. This can, of course, take many different forms, from a robot arm picking items in a warehouse to a humanoid robot trying to navigate 3D space in the real world.
When viewed in the context of current developments in artificial intelligence, where the AI can be thought of as the ‘brain’ responsible for processing information, robotics is concerned with physical interactions with the outside world, collecting information through sensory inputs and translating this information into physical outputs in the surrounding environment.
Robots and AI
A common mistake is to confuse robots with artificial intelligence. While it’s true that robots can contain artificial intelligence (sort of), they are actually entirely separate fields.
Similarly, the presence of AI isn’t a prerequisite to building a robot. Many applications of robotics focus on machines which follow computer programmed commands to complete a task and aren’t necessarily using any intelligence or autonomy.
Perhaps some of the best examples of these kinds of robots are those which have been designed to carry out highly-specialized, repetitive tasks on production lines or in warehouse environments. In these cases (until recently, at least), the need for any kind of automated intelligence wasn’t required. Individual robots were, for example, responsible for installing a seat on a car chassis in exactly the same pre-programmed way time after time.
Nevertheless, robots can incorporate artificial intelligence into the overall design of the machine and this is increasingly where much of the effort is now being focused.
When AI is integrated with robotics there is an increased level of autonomy with the robot which is now effectively able to analyze and evaluate the best course of action given the data provided from the outside world, either communicated from a third party source or collected from the machines itself using sensors.
The pairing of artificial intelligence with advanced robotics is ultimately the foundation of many current developments happening in both fields, with the application of technologies such as computer vision and natural language processing allowing for huge advances in everything from autonomous vehicles to human-like androids.
How Robots Work
As a discipline, robotics is concerned with programming machines to manipulate the physical environment. This can look like anything from walking down the street to carrying out dangerous tasks in areas too hazardous for humans.
When it comes to modern robotics, the tasks carried out by robots are almost always the result of a series of interactions between inputs gathered from sensors, outputs delivered by motors and actuators, and some kind of computational device which is responsible for processing, analyzing, and acting on all of the information gathered from the external environment.
Robots are basically machines designed to accomplish a task. Often (but not always) this is done through programming.
There are three main components when it comes to getting a robot to successfully carry out a task – whether that’s vacuuming your living room or driving a car safely from A to B.
- Sensors (for input information)
- Control Systems (for decision making)
- End Effectors and Actuators (for output actions)
While at first glance these don’t seem too difficult, on an individual level each of these areas are really challenging for a machine to carry out and even more so if the robot has to carry out the task autonomously and without human input.
When any intelligent or autonomous machine is brought into the physical realm the question of ethics becomes paramount. As with the development of artificial general intelligence which would see AI at the same level of cognitive intelligence as humans, the priority in developing advanced robotics is ensuring that robots are designed and programmed with values that are aligned with ours.
Science fiction writer Isaac Asimov understood the potential dangers of combining artificial intelligence with robotics when he introduced the fictional ‘Three Laws of Robotics’ in his 1942 short story ‘runaround’ which set out a trinity of guiding principles that he believed should inform the ethical programming of robots and intelligent machines.
Asimov’s Laws of Robotics
0. A robot may not harm humanity, or, by inaction, allow humanity to come to harm.
1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws
Although far from comprehensive, particularly when it comes to the some of the ethical considerations that will be prompted by robots of the very near future, Asimov’s Laws were the first such set of guidelines actively exploring this issue and have acted as a foundation for philosophers and researchers since they were first proposed.
Types of Robot
While the major unifying factor of robotics is that it refers to the autonomous physical presence of a computer program able to act independently in the real world, the range of ways in which this goal manifests is incredibly diverse.
From medical robots increasingly assisting (and eventually perhaps carrying out) complex surgical procedures, to airborne military drones, road-based automated vehicles, and even the push to create human-like machines, the variety of robots is increasing by the day.
Androids and Bipedal Robots
Simply defined as robots that look and act like humans, androids are the anthropomorphic projection of ourselves and are the ones most often associated with the ‘typical’ robot. A lot of work and funding is going into the development of androids right now, including getting robots to move and look like humans. When it comes to humanoid movement Boston Dynamics is leading the way. Their Atlas robot, for instance, is now able to carry out a realistic and somewhat terrifying demonstration of human-like movement in the real world.
At the same time, researchers have made significant progress in creating relatable robots that look like humans. For example, Hanson Robotics’ robot Sophia looks remarkably human-like. The hope is that this will help to create positive and meaningful relationships between humans and robots.
Navigating the Uncanny Valley
A major issue when it comes to creating humanoid robots is removing the distinctly un-human characteristics. The term ‘uncanny valley’ is used to describe the phenomenon of an artificial copy or digital representation of a human that just doesn’t seem quite right. Breakthroughs have been occuring on this front, as well. Take Saya, for example, a digital project by Japanese cg artists Teruyuki and Yuka Ishikawa which does a pretty convincing job of overcoming the uncanny valley.
Nevertheless, one big question still remains: how hard will it be to translate this realism into the real world, and if we ever manage it, will we still notice differences?
As with most engineering problems technological advances make it likely that we’ll see significant progress in the area of humanoid robotics in the coming years. When the traditional hurdles are gradually overcome it’s possible that we could get much closer to Westworld and Ex Machina style androids than we think possible today.
Quadrupedal and Animal-Inspired Robots
While our default is to think about humanoid robots, in many cases, some of the most advanced developments in the field of robotics are actually around animal-inspired robots. Four-legged robots have seen some of the greatest progress in the field of robotics research, offering superior stability and flexibility in many roles where four legs > two legs.
Again, companies like Boston Dynamics are currently leading the way and with almost unbelievable progress in the mobility and dexterity of robots such as SpotMini, it’s not hard to see why.
Another type of robot getting a lot of press attention lately is the autonomous vehicle, also known as self-driving cars. From the robotics perspective, the requirements are relatively straight: a self-driving car must deal with directions (turn left, right, go straight ahead) and speed (faster, slower, stop). The complication lies in the constant amount of sensory input data required to monitor every possible change and potential hazard surrounding the car. This includes lane monitoring, pedestrians, other vehicles, weather, and road signs.
In order to deal with the sheer complexity of navigating an always-changing environment, self-driving vehicles utilize complex computer vision algorithms to ensure they’re fully ‘aware’ of what’s going on in the world around them.
Self-driving cars are no longer the stuff of sci-fi and with major automotive companies developing the technology for roll-out as we speak, it’s likely that you’ll be seeing increasingly more autonomous vehicles on the road over the next few years.
Some of the more common kinds of robot in military use today include those used for bomb disposal and aerial reconnaissance drones. In both cases human operators maintain a central role in their operation. This is likely to change over time, however, particularly given the amount of funding provided by governments to research military applications of new and developing technologies and robotics.
In robotics, the next step in the field of military robo-integration are what are known as Lethal Autonomous Weapons which, as the name grimly suggests, are intended to operate on their own, and eventually to act independent of a human operator. This means that the final say on whether any action is taken will be done by the robot, rather than a human.
As with many other areas in the development of AI and advanced robotics, there are significant ethical issues around the responsible application of autonomous machines in these kinds of roles. From the larger moral considerations of weaponized robots to concerns around oversight and safeguards, there are many troubling questions which need to be asked about developments in this area.
The potential threats of weaponized robots has led to pioneers and leaders in robotics and AI, as well as politicians from at least 19 different countries, to call for an urgent ban on the development and use of killer robots.
Manufacturing and Warehouse Robots
Unlike humans, robots are predictable and don’t get tired. They can carry out pre-programmed repetitive tasks much more reliably than we can which makes them a natural addition to industries that require this type of labor. This is good news for robot manufacturers and companies that want tasks carried out efficiently and cost effectively and bad news for workers.
Vehicle production lines have long been associated with fleets of robotic arms working in sequence to carry out specific tasks along the assembly line. In many car production facilities today the role of the organic worker (humans) has effectively been reduced to maintenance, quality assurance, and general oversight tasks as incredibly sophisticated and highly specialized robot workers tirelessly install, weld, and paint at a pace and degree of consistency that humans simply can’t match.
With over 3 million industrial robots in factories around the world (a number likely to skyrocket in the coming years), goods are going to increasingly become cheaper to produce, store, and handle as manufacturing and distribution robots become more advanced and commonplace.
Scientific and Medical Robots
In the field of medicine, robotic assistance with complex precision tasks such as surgery has been around for a few of years now. One of the most well known examples of this is the da Vinci surgical system which has assisted with over 6 million operations worldwide since the early 2000s.
While far from autonomous, the current generation of robot-assisted surgery systems have transformed surgery. Using remotely controlled manipulators to greatly reduce the traditional problems of minimally invasive procedures, robot-assisted surgery systems improve the ability to carry out difficult and precision-heavy tasks.
Again, as technology in the field of AI develops alongside an ever-growing repository of medical data, it’s likely that we will see an increasing level of autonomy in medical robots. Not only in surgery, but perhaps even in other roles around hospitals and medical labs.
Probably one of the most obvious examples of a dangerous environment is space, which is quickly becoming one of the most ambitious proving grounds for robotic helpers. A great example of developments in the space industry is NASA’s Valkyrie, a bipedal humanoid robot designed to operate in the unrelenting and unforgiving conditions of space. Unlike pre-programmed or remotely manipulated robotic systems, Valkyrie incorporates artificial intelligence in order to work alongside, or independently of, humans also involved in future space missions.
Robots and the Future of Work
The ability for movement is what ultimately gives robots their power and even from the earliest days of robotics, it was this simple trait that made the prospect of ever-smarter and more capable machines a very real threat to human jobs. Of course, anxiety about increasing automation and worker displacement isn’t a new one. Since the industrial revolution of the 19th century, similar fears have existed in the face of technological advancements that drastically reduce requirements for manpower.
Whether the level of current fear and anxiety around job losses to robot workers is a valid one is still yet to be seen, however while we can’t say with any certainty just how significantly the rise of robotics will shake up the current work environment, it’s absolutely inevitable that it will. If the economics work, every job that’s possible to be automated almost certainly will be. It’s unlikely that many industries will remain untouched by robotic developments as increasingly intelligent machines begin to take over the role of everything from truck, bus, and taxi driver, to courier, warehouse staff, and even fast food workers.
The real question is whether new jobs to cater to needs that didn’t exit in the pre-robot age will spring up in the place of the old ones.
We’ll almost certainly find out the answer to this question within the next decade or so 🤖