From comics to mechanical exoskeletons of machines

At the headquarters of Manhattan Magic Comics Company, Tom Braive, editor-in-chief of Iron Man Comics, briefly introduced me to Tony Stark: a graduate of Massachusetts Institute of Technology, a star scientist/engineer, a billionaire heir, a playboy and an alcoholic. Later, the bad guys kidnapped Stark and coerced him to make deadly weapons for them. So he built a metal armor. After escaping, he suddenly woke up, improved his armor and turned himself from a selfish genius into a real superhero.

Iron man is faster than a jet plane, and can lift a weight of 1000 tons and break into a heavily guarded computer. Like a fable? Of course, Braive said, this is the charm of fantasy. "Iron Man must always be ahead of reality, otherwise he will become a fossil according to the standards of comics."

1963, when this role first appeared, the army was also conceiving its own "Iron Man". In the same year, Sergei ZaRudney, a weapon researcher in the US Army, published a report describing the wearable machine jacket he designed, which would give the wearer the power of Hulk, but the technology to realize this idea did not exist at that time. Except for some non-military designs, the real super coat has little prospect. Until 2000, Darpa began a seven-year, $75 million mechanical exoskeleton research program. At that time, a few supporters of mechanical exoskeleton thought that technology-including US Army Colonel Jack Obsuke-finally caught up with this idea. Obsuke has been helping to promote the research of exoskeleton since 1995. He said that as sensors become smaller and smaller, with more functions and faster processors, he and other supporters have reason to believe that mechanical exoskeletons may become a reality.

But Darpa's ambitious goals are unrealistic for anyone. It wants a magical machine that allows soldiers to drag things weighing hundreds of pounds for several days without feeling tired; It allows soldiers to flexibly operate weapons that usually require two people to control; You can easily leave the battlefield with two injured comrades on your back. They demanded that this mechanical coat should be equipped with armor, so that the enemy's firepower would be powerless. They even hope it can help soldiers jump higher. Anyway, they want the Iron Man in the comic book.

Before starting the plan, some consultants of Darpa immediately pointed out that their ideas were unrealistic. EveRacine Garcia, an engineer at Cornwall University who was in charge of the Darpa Iron Man project in the early days, said, "Half of the people I asked believed it, while the other half thought it was a waste of time, money and resources." There is nothing wrong with those who pour cold water, he added. "This is really a daunting challenge." The mechanical exoskeleton will need a photodynamic system, which can provide electric energy for several days; It also needs small and powerful artificial muscles; Complex motion control system. It must also be agile and responsive.

The exoskeleton must become a soldier's mechanical shadow, which can read his every move and imitate his every move in time. Even millisecond hesitation will cause a burden and make soldiers feel as laborious as walking in water. The sensor of this kind of mechanical coat must be able to read every slight movement applied to its whole body at the speed of thousands of times per second, and its microprocessor must be powerful enough to convert these data into instructions in time and send them to the mechanical limb to coordinate with the movements of the inner wearer.

Genius like Tony Stark is needed to solve these problems and find out how to put this system into machines with speed, agility, strength and endurance. However, this person is not a weapon designer, but a person who makes mechanical dinosaurs as a profession. Steve Jacobson's resume is rich and colorful. In the past 35 years, his works include an 80-ton robot dinosaur and the fountain of Bellagio Casino. However, he himself looks more like an amiable professor than a scientific madman. He is tall, with broad shoulders, straight back and gray hair. Before introducing XOS, he showed me around what he called the "terror tunnel". From the outside, it is easy to be regarded as a dentist's office, but this cave-like room is actually the headquarters of Sacos and the research and development department of the Engineering College of the University of Utah. He often teaches here. Although he used to make robots for many famous picky customers-he hinted that Disney was as harsh as the American army-Jacobson was still an academic researcher at heart. He called his thoughts friends to play with. He seems to care more about the solution of the problem than the final application of the solution. After passing a humanoid robot that can play table tennis, he stopped in front of a pair of mechanical hornbills that can sing. They were built for a local hotel. The difficulty is to make their movements lifelike, just like real birds. He said, "We only take the jobs we want to do because they can arouse our interest."

He talked about Kan Kan, and the topic changed from engineering to energy-saving biological systems (carrot-driven people! ) His talkativeness may be misleading. In fact, he is a secret-loving person, rarely interviewed by the media, and even refuses to disclose his age. During the visit, he pointed to a small unmanned ground vehicle (the new design of exoskeleton legs) and excitedly explained it. Finally, he asked me not to mention it in the article. I think his worry may be because many projects are funded by the army. On the other hand, he is also like a magician who doesn't want to reveal too many secrets.

The research projects of SACO-including prostheses and nano-motors-seem to be varied and irregular. However, Yves Racine Garcia said that it was because of his wide interests that Jacobson became an ideal candidate to accept the challenge of exoskeleton. He has proved his ability in software and mechanical engineering, and what is more rare is that he can constantly invent new things as needed. "He can design his own transmission; Able to design control system; Can design a machine and all its components. " Garcia said. Designing XOS absolutely needs such an all-round genius.

"When designing something like an exoskeleton," Jacobson said, "there are 25 subsystems that can be completed before moving on to the next step. Although the two main design goals are strength and endurance, it must also be able to do 75 different things. " Among all the robots he designed, XOS is obviously his favorite work, because it has the biggest challenge and the most problems, and he regards it as his "son". "There is nothing more ambitious than this. Others do not need such a complete and independent system, and do not need to achieve such strength, speed, endurance and agility. In 2000, SACO applied for investment from Darpa. Jacobson thinks he has found the answer to the problem of Darpa reward. The problem is how to make the operator interact with the robot. To prove his intuition, Jacobson asked company photographer Jon Price and his daughter to help him do an experiment.

In this experiment, Price plays the exoskeleton and his daughter plays the internal operator. With her back to her father, she stood on his feet with her toes pressed on his toes. They hold hands to help keep balance. She began to move forward. Price's task is to keep pace with her daughter and keep her feet under her feet. After a few minutes, they acted like a person. His daughter is completely in control-how fast she walks and when she turns-and Price just tries to imitate her step by step.

The demonstration proved to Jacobson that a clever machine only needs a few contact points, such as feet and hands, to understand the operator's action intention tied to it and cooperate with it. Although the theory is simple, it is quite difficult to practice. In the process of completing XOS, Jacobson and his team redesigned the micro-transmission device, improved the pressure sensor, invented a more efficient hydraulic valve, and even designed the aluminum pedal of the robot. The control system called "clearing the road" by the designer is the key to integrate all the components, and it is also the core to upgrade another robot he designed to "superman coat". Obsuke, who personally tried it on XOS, completely agreed: "A little load makes people tired easily." However, the control system of XOS reduces the burden on human body to near zero.

It is this control system that enables the demo operator jameson to lift the 200-pound barbell 50 times in a row, but his heart rate has not increased. When he pulled the barbell off the shelf, the sensor in his hand immediately detected the change of torque. Without the help of the exoskeleton, the sensor will show that he tried to pull down 100 pounds with each hand. However, Jacobson explained that the goal of the system is to make the pressure felt by these sensors close to zero, so that XOS can work.

The sensors on those hands transmit the measured data to the central processor at the speed of hundreds or even thousands of times per second. This system inputs data into a series of formulas to calculate the direction and movement of exoskeleton arms, legs and back. Finally, I realized that jameson wanted to put his hand down and figured out how each artificial muscle in each joint needed to move to imitate his movements. Jameson never felt any burden, because the system had commanded the mechanical arm to remove the barbell for him before he really exerted his strength. After the exercise, he unloaded XOS and showed no signs of panting. I asked him how he felt. "Good," he said and shrugged his shoulders.

The XOS demonstrated by Jameson is about the fourth edition. Jacobson showed me around a room, and the first three sets of prototype mechanical coats hung on the shelves like puppets. This immediately reminded me of Iron Man's "Armor Room"-the room where Stark kept his armor. The first mechanical garment was made in 2002, and there was no power. Sacos's team built it to prove that the exoskeleton can move freely like a human body. Jacobson tied an engineer to his clothes and asked him to try all kinds of complicated movements, such as kicking a ball, running and climbing into the cab of a car. Through this series of experiments, they confirmed that the right joints were used in the right places.

It is more difficult to open and close these joints with proper speed and strength. In 2003, Sacos began to manufacture artificial muscles with hydraulic actuators. This method was not initiated by him. In fact, another exoskeleton researcher believes that XOS's dependence on hydraulic devices will eventually lead to its failure. The engineer, who asked not to be named, has never seen XOS with his own eyes, but he has seen a news clip on YouTube. He said that the hydraulic device consumes too much electricity. In his view, the electric actuator is better because its energy consumption is consistent with the action. However, Jacobson impatiently refuted this criticism. "Do you like your car brakes? Do you want the landing system of the plane to work properly? They are all hydraulic. " Later, he added that he had solved the problem of energy waste. But he didn't want to explain the details, just saying that Sacos redesigned the valve to control the flow of liquid, so that it only started when necessary, so it would consume energy when the mechanical clothes were moving.

Although impressive in weightlifting, XOS has not achieved all Darpa's goals. It can't make you dunk in the air, it can't help you run faster, and it can't turn you into a strongman. However, Obsek said that one of Darpa's initial goals was to see if its wish list could be realized. Among the three teams involved in the project (Sacos, Oakridge National Laboratory and University of California, Berkeley), XOS stood out in 2005, which is the closest to the concept of the Pentagon and became the whole-body exoskeleton approved by the US military to enter the next stage of research. SACO has received $654.38 million from the military, covering two years of research.

Jameson put on XOS again to demonstrate relaxation exercises. I watched this 150-pound robot coat imitate his every move like a shadow, and only six joints were touching each other. Imagine the huge amount of data flowing between sensors and CPU every second. The scene in front of you is as wonderful as the special effects in Iron Man movies. I can hardly help but imagine jameson flying over the roof. However, this is unlikely to happen. To get out of the cave, you must first cut off the wire connecting XOS. The world of exoskeleton researchers is small, secret and full of intrigue. Although you may not understand the working principle of your opponent's device, you will still lose no time in cynicism. The most common way to attack is, "You ask him how he intends to provide motivation." XOS and other two sets of leading exoskeletons in American laboratories try to solve this problem from different angles. Jacobson decided to build a fairly capable spacesuit first, and then study how to provide it with electricity for 4 to 24 hours (the minimum power requirement proposed by Darpa). In all the demonstrations I have seen, jameson and XOS are connected to a hydraulic pump, through which they can get electric energy from an external power supply. This suit can be powered by batteries, but it can only work for 40 minutes at a time. Two other exoskeleton researchers-Professor Hugh Herr from MIT and Professor Hamayong Kazrouni from the University of California, Berkeley-have begun to solve the problem of motivation.

Hal is trying to build a foot-operated machine, which saves energy as much as possible-it only needs 2 watts to start, which is equivalent to the energy consumption of a portable radio-but it can support 80% of the wearer's 80 pounds. Because the design will affect the gait of the wearer, it will consume a little more energy after wearing it than when it is not worn. However, Hal believes that in the near future, he can improve the mechanical structure so that the machine will eventually help save rather than consume the wearer's physical strength. Finally, he imagined that the equipment would be used for entertainment. On weekends, people will wear it to climb mountains without feeling tired after running for a day. If Hull's vision is still far away, Kazrouni hinted to me that he had reached the halfway point to solve his motivation.

Kazrouni said that his "HULC" lower limb exoskeleton can work continuously for 20 hours. It enables the wearer to carry 100 pounds, but consumes less oxygen 15%.

Kazrouni's device cannot be demonstrated yet. He only revealed that the principle of the system is similar to that of a hybrid car. Hybrid electric vehicles convert the energy generated by braking into battery charging. HULC uses the energy returned from the ground when pedestrians change their feet. The action of walking itself is enough to generate a steady stream of kinetic energy. He has obtained $2 million from the National Institute of Standards and Technology to improve the system. In the end, HULC will help people with mobility difficulties to walk again. "This is not a war machine," he said. "Our machines may replace wheelchairs."

XOS's strongest competitor is also medical equipment, but it is located in Japan on the other side of the Pacific Ocean. In 2004, Japanese robot expert Yoshiyuki Shanhai founded a company called Cyberdyne (the same name as the company that triggered the robot revolution in the movie Terminator) to promote his whole-body mechanical exoskeleton HAL-5. Instead of using XOS pressure sensor, it attaches the sensor to the wearer and receives his or her muscle signals to determine his or her action intention. The control system of this mechanical coat can learn and imitate the natural posture of the wearer. This means that it takes at least 30 minutes to coordinate the two-you can't expect to move freely as soon as you put them on. However, HAL-5 is mainly used as an auxiliary tool for rehabilitation treatment and a nurse's assistant, so the training time of half an hour is not a problem. After wearing this battery-driven power suit, nurses can easily hold this stout patient like a baby. Shanhai Kaji has started renting HAL-5 to customers. In the Iron Man cartoon, the superhero is knocked to the ground and lies on the floor of the enemy's lair. The monitor in the helmet told him the bad news-the electricity was used up. But there is still hope. He inserted his finger into the concrete floor, found a wire and quickly completed charging.

Unfortunately, in the real world, charging the exoskeleton is much more troublesome. Therefore, the first XOS on the battlefield may even be connected by cable. Obsuke imagined that the original version was more likely to be an engineer than a soldier. After being connected to the power supply of warships or military vehicles, XOS can help soldiers quickly unload from helicopters loaded with heavy weapons or repair tanks with broken tracks. Although the US Army hopes to test the wired version of XOS on the battlefield before 2009, Jacobson and his colleagues are still nervously developing their own power version.

This summer, SACO will cooperate with an engine design company to develop an engine that can continuously supply power to XOS for several hours. In addition, Jacobson is reluctant to disclose more. He prefers to talk about another more interesting challenge-instead of building a powerful engine, it is better to reduce the energy appetite of XOS.

Jacobson showed me a new type of energy-saving mechanical leg, which imitated the design of human leg drive. When walking, our hips generate the most energy. When the leg is in front, the small muscles of the knee and its parts are completely relaxed, ensuring that our feet fall to the ideal position on the ground. This free swing technology is quite energy-saving. Kazrouni and Hal have designed it into their lower limb exoskeleton. Jacobson is designing it as a future version of XOS. "The next step," he said, may be a few years later, "will achieve the goal of walking at 1 to 3 horsepower." At this time, only one portable battery pack is needed to provide lasting power.

Jacobson regards today's version as a basic model, and eventually it will be transformed into various versions to perform different tasks, whether in hospital or on the battlefield. Future models may even be fully automated. "You go out and say to it,' Go to that building yourself, because I'm too lazy to go.' "

Later, when I walked through the hall of SACO, I saw a cartoon depicting the future XOS playing on the flat TV. In one clip, soldiers wearing XOS fly over a high wall with missiles on their shoulders and can even do elegant backflips. Although wrapped in mechanical bones, they look as flexible as football wingers. They look like Iron Man.