Passive dynamics is an approach to robotics which uses the momentum of swinging limbs for greater efficiency. A purely passive dynamic robot requires no power at all. It will able down a plank all by itself if it’s configured just right. Of course, in the real world robots can’t be expected to always walk downhill. Ideally, one could combine the efficiency of passive dynamics with some kind of power so it can walk efficiently on a flat surface.

In 2003 at the University of Sussex, Eric Vaughan used artificial evolution to create powered bipedal walkers in simulation. Evolving both the passive dynamic body and the neural control system all at once didn’t work, so the evolution was given some assistance. First, the robot was evolved to walk down an incline passively. As the evolution progressed, the plank was gradually lowered until it was completely horizontal, gradually bringing the neural control system into play. The final result was a robot design that could walk on a flat surface using very little energy, much like a human.

More:

• TU Delft Robot Denise 2 - passive pneumatic robot

In 1970, noting the extreme cost of space exploration, the physicist Freeman Dyson proposed a wild idea: send a machine into space that is capable of building copies of itself from materials it finds. This would provide an unlimited production and exploration capacity for a finite cost.

Although just a thought experiment, Dyson’s idea is not as crazy as it sounds. In the early 1980s, NASA funded a series of investigations into cheap space colonization which involved building self-replicating factories on the moon. It’s theoretically possible, but robot factories building more robot factories poses a huge maintenance problem. After all, you’d need an army of repair robots to fix things when they break down. And, who repairs the repair robots?

Eric Drexler, a proponent of nanotechnology, describes a more elegant solution. He envisions tiny, molecular “assemblers” that can build copies of themselves and other items of greater complexity — a bottom-up version of the self-replicating factory idea. He also notes that such technology, if not tamed properly, could replicate out of control consuming all the resources on the planet resulting in the so called Grey Goo Scenario.

 

Copyright Shane Willis

Thought experiments aside, some researchers have succeeded in building primitive, yet practical self-replicating machines. Adrian Bowyer from the University of Bath developed a rapid prototyping machine called RepRap that can make most of the parts necessary to build… another RepRap machine.

Bowyer’s RepRap project is now replicating “in the wild”. On November 30, 2008, the first user outside the lab used one of the machines to produce and sell a set of RepRap parts to someone else. Being a prototyping machine, RepRap is not limited to copying itself. It has been used to create ordinary objects including a coathook, a pair of sandals, and a fly swatter. The potential for such a machine is vast and is fortunately not likely to turn the world into goo anytime soon.

More:

• The RepRap Project
• Video: Cornell University’s Self-Replicating Robot Blocks
• Von Neumann’s Universal Constructor
  

How does the brain learn? Understanding that process would allow one to write software that could learn the same way humans do. But, prior to 1949, no one had a very good answer. It had already been well established that the functional unit of the brain was the neuron, and the structure of these neuronal cells had been studied extensively. However, psychologists did not have a good theory about how neurons produced human behavior.

Enter Donald Hebb, a Canadian Psychologist who was fascinated by how the brain worked. Hebb postulated that neurons form cell assemblies, collections of neurons that act in concert to produce behavior. This idea formed the beginnings of the field of connectionism, an approach to the mind that views complex behavior as emerging from an interconnected network of simpler units. But how do these networks form? To answer this, Hebb proposed a mechanism which has come to be known as “Hebbian Learning.” The idea stated simply is: “Neurons that fire together, wire together.”

The brain is fully connected at birth, but the strength of these connections changes through time as we learn, forming the cell assemblies that Hebb theorized were responsible for behavior. Hebbian learning postulates that when neuron A activates, and then causes neuron B to activate, then the connection strength between the two neurons is increased, and it will be easier for A to activate B in the future. The idea sounds simple, but it goes a long way in explaining how neural networks form in the brain. Not every learning process in the brain can be explained by Hebbian learning, but it does provide an explanation of how complex networks of neurons could form.

After Hebbian learning made its debut in the 1949 book, “The Organization of Behavior,” it then became possible to program computers with the Hebbian rule, giving them the ability to learn. Today, many different types of artificial neural networks (ANNs) are used extensively in the field of artificial intelligence, including applications in face identification, speech and handwriting recognition, financial applications, data mining, and even autonomous vehicles. Hebb’s discovery spawned a whole branch of artificial intelligence and methods for constructing learning mechanisms on computers. ANNs are not yet sufficient for creating human-level intelligence on a computer though. Real neurons are complex biochemical engines, whose behavior can only be approximated with ANNs. Also, human brains come pre-configured to some degree, and without understanding this innate structure well, building large artificial neural networks is not practical.

More:

• Donald O. Hebb

When building a robot that emulates an existing animal or other life form, engineers must accept the humble truth: nature always does it better. You just can’t compete with billions of years of evolution. Consider the agility of a simple house fly, or the fact that certain moths can identify pheromones when there are only a few molecules present. Some researchers have therefore concluded that it is far more effective to try and interface with some of nature’s technology instead of building it from scratch.

Jose Delgado pioneered the research into electrical brain stimulation at Yale in the 1940s by inserting electrodes into cats and monkeys to evoke responses in the brain. In a famous media stunt in 1963, he fitted a bull with a remote control device which stimulated a specific part of its brain. He stepped into the ring with the bull and let it charge him. At the last second, he activated the stimulator, stopping the bull in its tracks.

Since then, the military has gained interest in this technology as it could be used on insects for surveillance purposes. Various DARPA funded projects have succeeded in controlling insects and beetles, getting their wings to flap at different speeds based on a computer controlled electrodes inserted into specific parts of their brains. The big problem is trying to design components small enough that can be inserted into an insect’s body without disrupting its flight.

Chuck Higgins from the University of Arizona thinks a better approach is to turn the problem inside out by embedding insects into the robot. If a robot could leverage the chemical sniffing ability of a moth, for example, then it could be used by the military for detecting explosives. Higgins and his team have been able to build a robot that can read visual information directly from its passenger Hawk Moth’s brain using electrodes and amplifiers and send this information back into the robot’s control system. In the future, he’d like to integrate with other parts of the moth brain, including the olfactory system which would give his robots one of the most sophisticated sensory systems available.

More:

• Jose Delgado’s Bull Stunt
• DARPA’s Hyrbid Insect Micro-Electro-Mechanical Systems project
• Video - Chuck Higgins: The Brain-Machine Interface
• The Higgins Lab

In the West, robots are not to be trusted. After all, they might someday become super-intelligent and kill everyone. The Czech writer Karel Capek laid it all out for us in his play “Rossum’s Universal Robots” (1921), the first publication to use the word “robot.” In the play, robots are used for slave labor, but eventually stage a rebellion and destroy humanity. That basic theme has remained in Western sci-fi literature and movies ever since.

Not so in Japan, however. The Japanese are absolutely head-over-heels in love with robots. They are viewed as saviors, not destroyers. The cultural icon, “Astro Boy,” sums up the Japanese attitude well. Astro Boy was a wildly successful comic started after WWII that eventually became an animated series that endured for decades. It tells the adventures of a cute and beneficent android with incredible powers. He is brave, gentle, and wise, protecting humans from danger including alien invaders, robots gone berserk, and even robot-hating humans.

The development of robot technologies in Japan, funded in large part by the government, is focused on human-robot interaction, or social robotics. Big projects include robot receptionists, household servants, nurses and companions. While Americans are content to just switch on their robot vacuum cleaners and leave them be, the Japanese long for ongoing interaction. Paro, for example, is a cuddly model resembling a seal. Its purpose is therapeutic, providing comfort to the elderly and infirm. Only in Japan could such a conspicuously unemotional machine provide real long-term emotional comfort and companionship.

Japan’s robophilia can be partially explained by its demographics. Japan’s population has one of the highest average lifespans, but the country also has one of the lowest fertility rates. Soon, there will simply not be enough young workers to maintain the elderly population. Combine that with a healthy dose of xenophobia, and the most attractive option is to employ robots as the caretakers. The Japanese may be culturally primed for such a solution, because the native Shinto religion often blurs boundaries between the animate and inanimate.

Japan’s obsession with robots has made it the world leader in the field with twice as many industrial robots per worker as any other country. It has also created some of the most impressive demonstrations of advanced artificial intelligence, from Sony’s cute Aibo dog to the anthropomorphic servant, Asimo, built by Honda. There is still a long way to go with those technologies, but in the meantime, the Japanese are content to be surrounded by as many robots as possible. As their population ages, they look forward to having their mechanical friends and caretakers look after them to the very end.

More:

• Mitsubishi’s Wakamaru - household robot
• Paro - therapeutic robot
• Karakuri - 18th Century Japanese mechanized puppets
• Astro Boy (2009) teaser trailer

A “super-organism” is a group of organisms that act collectively to ensure their own survival. For example, a beehive consisting of thousands of individual bees can be considered a single organism. The hive has a life of its own, usually living 8-10 years, while the individual bees that comprise it live only 1-2 months each. A hive also exhibits division of labor, similar to the organs of an animal, where groups of bees are responsible for specific functions. Reproduction is even specialized within the hive, as the Queen is the only bee allowed to reproduce. Individual worker bees are therefore selfless, working only to ensure the survival of the Queen and her DNA. Like a single cell in your body, a worker bee’s own survival is trivial as compared to the reproductive process of the organism as a whole.

But, does a beehive have a collective mind? For an outside observer, it would appear so. For example, when new sources of nectar are discovered nearby, the entire hive can be rallied into activity, as more foragers are sent to the source, and more storer bees are recruited inside the hive to handle the influx. Also, when it’s time to move the nest, the hive considers optimal locations by sending out scouts. Then, once a suitable location is chosen, the entire hive is quickly relocated in an organized fashion. No single bee has the entire plan. In fact, each bee only has a tiny bit of information about the activities of the hive as a whole, including the Queen. A beehive is therefore a good example of Emergence, where complex behavior can result from the interactions of a set of relatively simpler behaviors.

In the case of beehives, the key to generating complex behavior is based on 1) the concentric organization of the hive, 2) the presence of environmental cues, and 3) bees’ ability to communicate with each other. Hives are organized from the Queen outwards, and this physical organization will dictate an individual bee’s career path. When a bee is born, it stays close to the Queen, grooming it and cleaning nearby cells. Then it can be recruited to storage tasks, taking incoming nectar, pollen, and water and storing it in the honeycomb. Finally, the bee can be recruited to a scouting or foraging role outside the hive. Within the hive, each bee has access to certain “global variables,” such as temperature and nectar throughput. This information, combined with the bees’ ability to communicate with each other through the various “recruitment” dances, results in the complex behavior we see.

Artificial intelligence researchers are interested in emergent behavior as this might be a viable means of creating complex systems that exhibit intelligent behavior. Like the hive, the human brain is comprised of smaller, simpler units, whose individual behavior is simpler to describe. Examples of emergence abound in nature, but also in human societies, economics in particular. No single trader can fully understand the nature of the stock market as a whole, but the collective actions of traders together result in a complex system capable of maintaining efficient prices, and sometimes acting quickly and collectively in response to new economic environments.

More:

• Emergence
• Bee dances: Waggle, Tremble, Grooming
• Eusociality

Joseph Ayers has a thing for lobsters. He likes to cook them, eat them, and build them. In his book, “Dr. Ayers Cooks With Cognac” he describes several creative lobster recipes from Cajun to Mediterranean and everything in between. His favorite way to eat a lobster is simply dipping the meat in butter accompanied with a glass of chardonnay. And, when it comes to building robot lobsters, he prefers the kind that can wander the sea floor searching for mines.

A biologist and neuroscientist, Dr. Ayers developed his robo-lobster at Northeastern University between 1999 and 2002 with funding from the Office of Naval Research (ONR) and the Defense Advanced Research Projects Agency (DARPA). Working from Northeastern’s Marine Science Center at East Point in Nahant, Massachusetts, Ayers looked to real-life lobsters for inspiration. He notes that lobsters are at the top of the food chain in their environment and so have a good built-in seek-and-destroy mechanism.

His robots are controlled with artificial neural networks, which are used to make subtle decisions such as whether to walk over or around a rock. When it comes to tweaking the design, Ayers always went back to the original, studying the behavior of living lobsters, trying to replicate their behavior. The plastic antennas sense obstacles, the eight legs can propel it in any direction, and the claws and tail keep it stable in turbulent water.

The robo-lobster’s actuators are controlled by Nitinol, otherwise known as “muscle wire”. Nitinol can change shape when a current is passed through it, making it more life-like and less “robotic”. Also, Ayers’ neural networks use Central Pattern Generators (CPGs), specialized neural circuits found in all animal brains that automatically generate a rhythmic patterns. The CPG acts as the source for the robots pattern of locomotion.

The robo-lobster is an excellent example of biomimicry, where ideas from nature are incorporated into robotic designs, however, a practical mission for the robo-lobster has yet to be demonstrated. In addition to finding mines, Ayers notes that it might also be used to measure pollution levels on the ocean floor. The robo-lobster was put on display at the Cooper-Hewitt museum in New York in 2007 as part of its “Design Life Now” exhibit. It was also named one of the Coolest Inventions of 2003 by Time Magazine.

More:

• Ayers’ recipe for Baked Stuffed Lobster
• Biomimetic Underwater Robot Program at Northeastern University
• The US Military’s Office of Naval Research (ONL)

Using artificial evolution to design robots is a powerful technique. The difficult part is writing a simulation of the robot and its physical environment. A simulation is required so tens of thousands of design variations can be automatically tested as the evolution progresses.

What if you didn’t need to write a simulation? Imagine the robot could figure out how to simulate itself. Not only have you saved yourself a lot of programming, but the robot is now very resilient. You could snap off one of its legs, for example, and it would be able to re-simulate itself, and re-evolve a walking strategy.

Hod Lipson from Cornell University’s Computational Synthesis Lab has built such a robotic system. His robot first squirms around on the floor, testing theories about how its own motors work and what its own shape is. This simulation of itself is then used to evolve a walking strategy.

Such a system would be very useful for planetary exploration, where robot engineers may not be readily available to fix problems. If the machine breaks a leg, no problem. Re-simulate, re-evolve, and move on.

More:

• Cornell Computational Synthesis Lab (CCSL)

If you plot all the important events that have occurred in the history of planet Earth on a chart, you’ll notice an interesting trend. The time between these events appears to get shorter and shorter as you get closer to the present day.

According to a group of scientists led by Ray Kurzweil, around the year 2045, the time between major technological advances will be so short that humanity will experience a cataclysm called the technological singularity. Machines will become more intelligent than humans and start to program themselves to become even more intelligent, creating a sort-of “intelligence explosion.” Nanotechnology will be completely integrated into our brains, enhancing us into a trans-human state. Kurzweil has even founded a non-profit institution, the Singularity Institute for Artificial Intelligence, to investigate the the impending phenomenon and gathers various scientists and futurists together each year at the Singularity Summit.

Naturally, there are many skeptics. The Pulitzer-Prize winning academic, Douglas Hofstadter, who regularly attends the Singularity Summit has said, “I don’t think it’s inconceivable that some kind of singularity entity could eventually have superior intelligence to humans, but I’d be very surprised if anything remotely like this happened in the next 100 to 200 years.” The futurist, Theodore Modis, has claimed that technological innovation is actually in decline. Kevin Kelly, the founding executive editor of Wired magazine, makes some interesting observations of Kurzweil’s charts of the singularity: according to the mathematical model, the singularity should be happening NOW, not in 40 years. Further, one could make the same statement at any point in history. During the industrial revolution, for example, one could have claimed that a singularity should have been happening THEN.

The major claim from the Singularity Institute is that it is not possible to make predictions about what will happen after the singularity occurs. Once humans “cross over,” we will forever be changed, integrated with technology, absorbed by it, or destroyed by it. No one can know. However, Kurzweil believes we will become immortal. Our minds will no longer be dependent on our bodies. We will be able to download, migrate, enhance, or repair our minds at will, forever living within the cradle of high technology.

This idea, along with the insistence that the world is fundamentally going to change on a specific future date, has caused some to point out similarities between the technological singularity and The Rapture, and so it has been called “The Rapture of the Geeks.” Kurzweil is considered a longevity expert and has written about it in his book “Fantastic Voyage: Live Long Enough to Live Forever.” The idea is, if you can just live just long enough to make it to the singularity, then you’ll live forever.

It should come as little surprise that the idea of the technological singularity is based heavily in science fiction. Victor Vinge wrote several sci-fi novels about the concept and first introduced the term “singularity” to describe it (taking it from the field of physics). Its a compelling idea, and the dystopian future of robots taking over is such a familiar theme that the idea of the singularity is easy to accept on the surface. It’s clear we live in a unique time period in the history of the planet. But, whether such publications should sit on the shelf of popular science or science fiction is still open to debate.

More:

• Video of Kurzweil’s talk at the Stanford Singularity Summit
• “How To Survive A Robot Uprising - Tips on Defending Yourself Against The Coming Rebellion” by Daniel H. Wilson
• Moore’s Law

In 1986, the German neuroscientist, Valentino Braitenberg, made an important observation about human psychology. He said, “When we analyze a mechanism, we tend to overestimate its complexity.”

Try following an ant around while it searches for food. You might infer a complex mind that is processing a myriad of difficult decisions. However, the ant’s actions can be explained by a very simple set of rules.

Even so, it’s not always easy to determine what those rules might be, looking in from the outside. Braitenberg therefore concluded that, when studying complex behavior, you must invent your own rules, from the inside looking out. In short, you tinker from the simplest possible starting point and see what emerges.

This line of thought led to his famous book, “Vehicles: Experiments in Synthetic Psychology,” which details a series of thought experiments in which simple mechanisms seemingly exhibit complex behavior.

Consider, for example, Braitenberg’s vehicle #2a. It’s a two-wheeled robot with light sensors. When light on the right side increases, so does the right wheel’s speed. Same with the left side. The result is a robot that avoids light. Cross the wires, and you have a robot that steers and accelerates towards a light source (vehicle #2b). While the simplest of parts comprise both robots, Braitenberg describes the former as “cowardly,” and the latter as “aggressive.”

How about robots that display “love?” Braitenberg’s vehicle #3 adds some inhibitory connections, which cause the robot to approach a light, and then gradually stop to stare at it indefinitely.

Other adjustments can create robots, which Braitenberg describes as “exploring, knowing, displaying instincts, making decisions, having foresight, ego and optimism.” Obviously, these are subjective, anthropomorphic terms, but Braitenberg purposely uses them to make his point: complex behavior can arise from simple mechanisms, IF they are configured just the right way.

More:

• Video of LEGO Braitenberg vehicle #2 from the University of Alberta’s Biological Computation Project
• A description of Braitenberg’s vehicles #1-14
• Herbert Simon’s Ant on a Beach