One kind of robot has endured for the last half-century: the hulking one-armed Goliaths that dominate industrial assembly lines.

These industrial robots have been task-specific -- built to spot weld, say, or add threads to the end of a pipe. They aren't sexy, but in the latter half of the 20th century they transformed industrial manufacturing and, with it, the low- and medium-skilled labor landscape in much of the U.S., Asia, and Europe.

You've probably been hearing a lot more about robots and robotics over the last couple years. That's because for the first time since the 1961 debut of GM's Unimate, regarded as the first industrial robot, the field is once again transforming world economies.

Only this time the impact is going to be broader. Much broader.

Today, robots are cropping up in offices, hospitals, and schools -- decidedly non-industrial environments -- as well as in warehouses, fulfillment centers, and small manufacturing centers. More and more, they are on our roads and flying overhead.

And that's just to name a few spheres in which robots are rapidly gaining traction by doing work more efficiently, reliably, and for less money than previously possible.

That's got a lot of people excited -- and a lot of others worried. The stunning pace of development in the industry has raised lots of questions.

This guide, written with the enterprise in mind, will address the big questions. And it'll give you the context to make up your mind about others. It'll also give you a handle on an industry that's poised to drive $135.4 billion in spending by 2019, one whose relevance to commerce and day-to-day life in the coming decades cannot be overstated.

Robotics geeks debate this over beers. No one wins. That's because any definition is bound to be arbitrarily rigid or too general.

Is your washing machine a robot? Is a modern high-end car, which engages in thousands of processes without the driver's knowledge? In truth, it's a little like Justice Potter Stewart's definition of pornography: You know a robot when you see one.

Need a better definition?
A robot is a programmable machine that physically interacts with the world around it and is capable of carrying out a complex series of actions autonomously or semi-autonomously.

There are four reasons:

falling sensor prices
open source development
rapid prototyping
convergence of disparate technologies
The demand for mobile computing has been a boon for robotics development, leading to falling prices, rapid advances, and miniaturization of sensor technology. Accelerometers used to cost hundreds of dollars each. Now every smartphone can measure acceleration, as well as capture stunning video, fix geographical location and offer guidance, interface with other devices, and transmit across several bands of spectrum -- functionality robots need to maneuver through our world productively.

The ubiquity of IoT devices is another driver. By 2025 there will be 100 billion Internet of Things connected devices generating revenue of $10 trillion. For the first time, sensors that capture and send data related to pressure, torque, and position are dirt cheap, leading to a boom in robotics development.

Similarly, prices for lidar and infrared sensors, previously the most expensive sensing equipment for self-guiding robots, have plummeted 90 percent thanks in large part to the aggressive development of self-driving cars by Google's Waymo and others. And 3D cameras, which used to be out of reach to all but the most lavishly-funded R&D teams and Hollywood titans, are now available off-the-shelf thanks to some smart work with algorithms.

Open Source Development

In 2009, a paper presented at the IEEE International Conference on Robotics and Automation (ICRA) introduced the Robotic Operating System (ROS) to the world. ROS is the first standard OS for robotics development. It also happens to be free, open source, and inherently flexible, freeing robotics developers from the time-prohibitive task of developing an OS from scratch.

There are plenty of open source users in personal computing, but because proprietary operating environments like Windows reached scale first, open source options have always been an alternative to something else. Not so with robotics, where open source is now the norm, resulting in a flurry of crowd-assisted development.

Open Robotics, under whose stewardship ROS falls, has also unveiled a robotics simulator called Gazebo which allows engineers to test robots in virtual reality without risking hardware.

How impactful have ROS and Gazebo been? Of the 23 teams competing in the vaunted DARPA Robotics Challenge, 18 utilized robots running on ROS and 14 used Gazebo to test their humanoid competitors in virtual environments.

The proof is in the investment. In 2015, more than $150 million in VC funding went to companies developing robots that run on ROS.

Rapid Prototyping

Though we're still waiting to see if 3D printers will fundamentally change how (and where) consumer goods are manufactured, the impact of additive manufacturing on robotics development has been enormous. "3D printing enables the creator to go from a mind-bending concept to a solid product in a matter of hours (or days)," according to Robotics Tomorrow, which tracks the industry.

Printers in maker spaces and university engineering departments, some of which allow for multi-material and metal printing, have significantly lowered the barrier to entry for robotics development. Need proof? Just check out the number of robotics projects that are live on Kickstarter right now.

When engineers can make prototype components at their workbench, innovation follows.

Technology Convergence

Just as its brought sensor prices plummeting, the enormous success of mobile computing has spurred advances in voice and object recognition, which have clear applications in robotics. 3D gaming sensors are helping robots navigate the clutter of the unstructured human world. And companies like Google, Amazon, and Apple have been hard at work bringing limited Artificial Intelligence platforms online and into homes.

This has all been accompanied by predictable year-over-year increases in computing power, along with the arrival of the cloud and IoT technology. Put it all together and you can see that a lot of technology that roboticists have been waiting for has matured in just the last few years.

Doom and gloom argument

That sound you hear? A big can of worms opening. Very smart people have staked out diametrically opposed views on this issue, and I advise extreme suspicion of anyone who speaks about these things with unnuanced certainty.

There are certainly some harbingers of bad news. A recent study by the National Bureau of Economic Research looked at the impact of increased usage of industrial robots on US local labor markets from 1990 to 2007 and found that there were "large and robust negative effects of robots on employment and wages across commuting zones." According to the historical data, jobs lost to robots have not been adequately replaced by new opportunities brought by robots, an argument technologists often fall back on.

Those findings are not predictive and should be taken in proper context -- the current boom in robotics largely started after 2007, and it's difficult to correlate the impact of robots on employment in industries as disparate as manufacturing and healthcare.

But the fears are real enough that heavy hitters are taking note. Bill Gates has voiced support for a robot tax, for instance -- a levy on the work robots do, which would replace income tax lost by the government when a robot takes human jobs. South Korea has come closest to that vision and appears ready to close tax incentives for companies investing in automation. South Korea's president is worried that higher unemployment in the robotic age will necessitate a robust welfare system, which is a huge problem since the government would be collecting less tax revenue to pad such a system during an employment crisis.

A recent report by Price Waterhouse Cooper suggests that up to 38 percent of US jobs could be lost to automation by the early 2030s. "The risks appear highest in sectors such as transportation and storage (56%), manufacturing (46%) and wholesale and retail (44%), but lower in sectors like health and social work (17%)."

But such findings are necessarily speculative, which accounts for the dramatic range of seemingly credible predictions about the future of employment once machines can do a lot of the stuff currently done by humans.

It's not that clear cut

On the other side of the debate, there's a credible argument that automation has resulted in regional job losses, but net job increases. One proponent of this view is the trade association A3, which released a study that found that during non-recessionary periods going back to 1996, both general employment and robot shipments increased. "To us," Jeff Burnstein, president of A3, told me, "that means that robots weren't killing jobs."

A few years ago, the International Federation of Robotics issued a study that looked at robotics use in China, Japan, Brazil, and India. As robot use accelerated in those countries, unemployment fell.

IDC recently found that spending on robotics will reach $135.4 billion by 2019, up from $71 billion two years ago. According to the report, services such as training, deployment, integration, and consulting will account for $32 billion of that, which accounts for a lot of new jobs.

Even the oft-cited PWC report isn't all doom and gloom. Robots increase productivity, and productivity gains tend to generate wealth. Historically, that's led to an increase in service sector jobs, which aren't easy to automate.

There are plenty of holes to poke in the methodology of all these reports. And that's the point: An accurate method for predicting how technologies will change the future is illusive -- and that's especially true when the technologies under consideration will fundamentally alter the economic paradigm. In the broad wake of that uncertainty, you have Ray Kurzweil predicting utopia and author Martin Ford predicting something much bleaker.

Ultimately, the PWC report comes to what may be the most sensible, albeit frustratingly vague, conclusion. It's not really clear what's going to happen. Average pre-tax incomes should rise with increases in productivity. But the benefits won't be spread evenly across income or education groups.

There are lots of categories to choose from, but you should know about these:

collaborative robots
telepresence robots
warehousing and logistics automation
healthcare robots
self-driving vehicles
Collaborative robots

A new generation of collaborative robots has emerged in the last few years. Unlike the heavy industrial robots of the 20th century, these collaborative bots, most of which have one or multiple articulated arms, are flexible and easily reprogrammable on the fly. Many models learn by watching humans demonstrate tasks.

The primary feature that makes collaborative robots from companies like Universal Robots, Rethink Robotics, and ABB safe is their ability to avoid unwanted collisions and, using high accuracy torque sensors, to recognize when they've bumped into something or someone they shouldn't have. That capability allows the bots to function outside of safety cages and alongside humans, which opens up new productivity potential for industrial manufacturers. The robots can learn complex tasks and then act as a second pair of dexterous hands to augment the capabilities of skilled workers -- thus the "collaborative" designation.

Why is it a game changer?
Automation is increasing in industries like automotive and electronics manufacturing and making speedy inroads in order fulfillment warehouses. As prices for task versatile platforms fall, small- and mid-sized manufacturers are starting to employ robots. Even so, a plausible future that sees robots replacing industrial workers entirely is far on the horizon, and in the meantime, with the economics favoring a hybrid approach, safety is of primary concern.

The market for collaborative robots could reach $3.3 billion by 2022.

Telepresence robots

Telepresence robots, which have been something of a novelty, are starting to creep into broader use. There are several different types, from the bare bones Double models, which are basically iPads on wheels, to iRobot's $30,000 Ava 500.

Why is it a game changer?

Across most sectors there's a growing segment of contract workers and freelancers who can't be in the office full time, and offices are seeing the value of poaching talent across time zones. Telepresence robots offer a surprisingly adequate alternative to being physically present. I've had a chance to try a few models, and the ability to navigate around the office really does differentiate the experience from a simple Skype call.

The market for telepresence robots could reach $8 billion by 2023.

Warehouse & Logistics

Of all the categories of robots covered here, warehouse and logistics automation is having the most substantial impact on global commerce right now.