ROBOTS: “Alpha Dog” – “BigDog” Taken to the Next Level

 15 May 2014

Alpha Dog

            As short a time as 15 years ago, it seemed almost impossible to imagine a walking robot.  At least, a walking robot that wasn’t an entertainment device.  Sure, you could design a device that went through all the motions of walking.  But it would walk on an ideally flat surface with no external physical interference or interaction of any kind.  In other words, most walking robots were not so different from those animatronic devices displayed in Disneyland shows.

            As long as the environment was carefully adjusted to the limitations of a walking robot, everything would be fine.  But that’s not what DARPA wanted.  The DARPA program required a robot that could . . . really walk.  This robot’s purpose was to accompany soldiers, potentially in combat situations, as they walked through rough terrain.  Just as humans and animals adjust their walking to the terrain, so would the robot envisioned by the DARPA project. 

            In other words, DARPA wanted a robot that could, and would, walk in every sense of the word. 

 



 

Big Dog

 

            The initial project, undertaken by Boston Dynamics, resulted in the unveiling of quadruped (four-legged) “Big Dog” in 2005.  It’s funny, but there’s something almost creepy looking about Big Dog in every still shot.  But when you see it move, the creepiness disappears as the viewer clearly recognizes something “familiar” and “natural” in walking motion of the robot.   Just watching Big Dog in motion wordlessly defines the term “biomimickry” – a technology copied from (imitating) nature. 
 

Big Dog (at the Beach?)

 

            LS3 is the“Legged Squad Support System.” And Big Dog was just the beginning. 

            What complex problem was this new “legged” technology designed to solve?  The Army identified “physical overburden” as major problem in warfare.  The modern soldier carries a substantial load of gear -- weighing as much as 100 pounds.  Both soldiers’ performance and readiness are impaired by the physical weight of their gear. 

            Well, “in the old days,” this problem was solved with a mule.   Accompanying soldiers, in the field, was a load bearing animal, a mule, which carried a lot of heavy gear leaving the soldiers less fatigued and more responsive to expected and unexpected challenges.

            In 2012, Boston Dynamics unveiled the LS3 -- “Alpha Dog” – Big Dog taken to the next level.   
 

Alpha Dog

            Alpha Dog can carry a bigger load – increased from 340 to 400 pounds.   This new version is quieter making a sound considerably quieter than the “swarm of bees sound” made by its predecessor.   While Big Dog was vulnerable to “cow tippers,” Alpha Dog and can “right” itself if tipped over. 

            Alpha Dog is also faster than its predecessor.  The robot manages a 1 to 3 mph walk over rough terrain and a 5 mph jog over relatively level surfaces.  On a flat surface, such as a roadway, Alpha Dog can reach a speed of 7 mph. 

Alpha Dogs

 

            Put in practical terms, the goal is to develop a robot that can travel with a squad of soldiers as they complete their mission – without hindering that mission in any way.  In order to do this, Alpha Dog will have to be able to follow the squad, but with a degree of independence or autonomy. 

            While Alpha Dog will respond to voice commands, the commander cannot command Alpha Dog in its every action without the robot becoming more of a burden and less of a help to its handler.  So, Alpha Dog’s design is must incorporate certain “autonomy settings.”   These settings will include: (1) “leader-follower tight,” (2) “leader-follower corridor,” and (3) “go-to-waypoint.”

            (1) Leader-follower tight: Requires the Alpha Dog to follow the leader’s path as closely as possible. 

            (2) Leader-follower corridor: Requires Alpha Dog to “follow” the leader, but with the “freedom to make local path decisions.”  So, the leader will not have to worry or account for Alpha Dog’s mobility capabilities.  The robot can vary its path slightly to avoid obstacles or obstructions without any special intervention from the leader.

            (3)  Go-to-waypoint: Requires Alpha Dog to proceed to particular set of GPS coordinates without a leader – avoiding obstacles on its own.

            A reasonable question: How can it do these things unless it can see?

            Well, for its own purposes, it can see.

            Alpha Dog has a “stereo” vision system.  First, it has a pair of cameras mounted into its “head.”  Second, each camera focuses on the same object or location from a slightly different angle – like human vision.  The angle to which each camera must adjust to focus on a distant object or location, discloses the depth, or distance, of that object or point.  

            But Alpha Dog also has a LIDAR detecting and ranging system.  LIDAR is a just a combination of the words “light” and “radar,” but is often assumed to be an acronym for “LIght Detection And Ranging.” (A useful factoid when you are trying to locate resources about this technology.)  Not only does the LIDAR system help Alpha Dog follow a human lead, but also records intelligence data directly from its environment.

            This type of sophisticated, simulated vision is necessary to allow Alpha Dog to meet another basic DARPA project requirement.  Without the “perception” capability to detect and judge both distance and grade, this robot wouldn’t be able to “walk” up and down hills.

            In terms of communication, Alpha Dog can’t give orders, but will be able to take orders.  Voice recognition technologies allow squad members direct spoken commands to which this robot responds.
 

Alpha Dog Takes Voice Commands
 

            It seems almost anticlimactic to add that Alpha Dog will be equipped with technologies to recharge batteries.  (It’s almost like saying the ‘bot can, also, open soup cans.)  But a mobile auxiliary power source is important to a squad in the field.  We are talking about batteries that power radios and other handheld technologies used by squad members on patrol.

            There will be “more.”  DARPA’s final goals for the perfected ‘bot will include a much larger load-carrying capacity.  The current 400 pound maximum will need to increase to 1,000 pounds – the weight of the gear required by a nine-man infantry squad on a 3 day mission.  Although Alpha Dog’s walking speed is about to par, its range will have to increase to allow the ‘bot to walk at about two-and-a-half mph for 8 hours.  Also, the ‘bot must be able to “burst” into 220 yard sprints at a speed of about 24 mph. 

Thursday 15 May 2014

GCLM5444HOxenia

See also: Big Dog -- Terrestrial Support Robotics

ROBOTS: BigDog – Terrestrial Support Robotics

15 May 2014

            Recently unmanned underwater drone technology seems have taken center stage among DARPA Programs.  For a while, proposals seemed to focus on UAV’s, unmanned aerial vehicles, but now include new, ambitious, underwater projects like the Hydra and UFP, which have recently appeared on the horizon.   

 Hydra UUV for the ocean's shallow waters

            On the other hand, Lockheed’s SSMS may be one of the first in new wave of terrestrial unmanned logistical and support vehicles.  The SSMS, Squad Mission Support System, vehicle sports the familiar wheels of most terrestrial vehicles.   But wheeled vehicles are of limited utility in many contexts.  I can’t help wondering when a new class of large terrestrial unmanned vehicles, with legs, will become the order of the day or, in terms of development proposals, the order to tomorrow.

Army, Lockheed to test drones-only mission, by air and land

            If the goal is high speed, accessibility, and maneuverability, terrestrial robots such as FastRunner (robo-ostrich) are prototypes intended to exploit to the maximum many of the advantages of bipedal locomotion. 

 FastRunner or "Robo-Ostrich"

            On the other hand, (or maybe on many other feet), there is a new generation of many-legged robots, most notably a group of hexapodal robots.  Most of these six-legged robots are designed more for the purposes of entertainment or amusement than military application.  But the multi-legged robot has distinct advantages over a two-legged or even four legged counterpart in terms of stability in motion over extremely difficult terrain.

            Looking back, from a mechanical standpoint, the business of walking was so complex that it seemed almost impossible to imagine a practical robot design incorporating motion – on foot – as short a time as 15 years ago.  But that changed with Big Dog.

 BigDog

            In 2002, Boston Dynamics [2] began work on a four-legged robot for military use.  Funded by the DARPA (Defense Advanced Research Projects Agency), the first prototype of this robotic quadruped was unveiled in 2005.  What had first been called, "Robo-Mule," but now renamed "Big Dog," had been developed by Boston Dynamics with Foster-Miller (a division of Qinetiq North America), Jet Propulsion Laboratory, and the Harvard University Concord Field Station. 

            Big Dog is about three feet long, two and a half feet high with a weight of 240 pounds.  In terms of size, it is roughly comparable to its inspiring model, the mule.  The DARPA program required a robotic pack animal, like the army mule, to travel "on foot" with soldiers through terrain too rough for wheeled vehicles.

            The latest prototype is capable of walking through terrain rough enough to stop a jeep.  Big Dog can run at about 4 mph with a 340 pound load and can climb a 35 degree grade. This robot carries a computer that receives feedback from the robot's sensors and controls its direction, movement, and balance.

            Powered by an “impressive” two-stroke, one-cylinder, 15-HP go-kart engine, Big Dog had a few bugs.  It could be tipped like a cow.  But unlike a cow, it couldn’t get back up.  Also, it was anything but silent -- making a sound often compared to a swarm of bees.  But since the 2005 unveiling, there has been a lot of work and refinements as well as the addition of a robotic arm that not only can pick things up, but throw them as well.  With the first unveiling, Big Dog's capabilties may have seemed modest, but this was the beginning of a new generation of walking robots inspired by biological organisms:  What's called biomimickry.

            In the 1950’s, the sci-fi vision of robotic technology was both exotic and strange.  The technology of the future was envisioned and presented as something completely different and contrary to our natural biological surroundings.  However, when technology confronted reality, we biological organisms seem to have had the last laugh because we could (and still can) do a whole lot of extremely useful things that our most sophisticated robotic technology cannot.

            The jeep took a basic automobile and raised the center of gravity, increased the size and scale of the automotive suspension system and produced spectacular off-road performance -- for a machine with wheels.  But the wheel, itself, was limiting.  Every Rover we’ve landed on Mars ended its life when it got stuck.  Human beings aren’t the strongest animal in the forest, but if just two of us were on Mars with those Rovers, we’d have extended their useful lives by getting them “un-stuck” in short order.  Why?  Because we have a repertoire of movements and leverage that we can use to apply force in almost any direction.  The best of those early sci-fi ’bots looked high-tech but, in fact, were functionally stunted.

            When sci-fi was still dominated by those inhuman and unnatural versions of mechanistic technology, a new methodology of approach to technological design was, quietly, born.  “Biomimetics” was a term used to describe the development of technology designed to imitate and replicate the activities of biological systems and organisms.  Then, the term “bionic” was coined to describe a technology incorporating a “function copied from nature.”  When Hollywood got a hold of the term “bionic,” the “Six Million Dollar Man” hit the small screen.  Maybe Hollywood’s version of the term “bionic” was just too interesting to be seriously “scientific,” and the term “bionic” fell into scientific oblivion.

            The gap was finally filled with the introduction of the term “biomimicry,” which has been widely adopted to describe any technology imitating (copied from) nature.  But, in some contexts, biomimicry is more of a necessity than a choice.  If you want robots or drone vehicles that work in a particular way, and the only known example of such performance is a biological organism, you’ll either have to imitate the organism or forget the project altogether.

            I am still amazed and entertained by the videos of Big Dog’s performance.  The movements are, in some ways, so “life-like” – so reminiscent of the movements of an animal.

 

BigDog (in Winter?)

 

Thursday 15 May 2014

GCLM5444HOxenia

See also: "Alpha Dog" -- "Big Dog" Taken to the Next Level

ROBOTS: What is “Cyro – the Robotic Jellyfish”?

15 May 2014

            Funded by the United States Navy, Virginia Tech has developed a “life-like, autonomous” underwater robot.  “Cyro the Jellyfish” is about the size of a human being with a weight of 170 pounds and a length of 5 feet 7 inches.

            Cyro has a waterproof shell attached to eight mechanical arms.  A sheet of pliable silicone is stretched over both the shell and arms.  The sheet flexes as the arms move giving the moving robot the distinct appearance, and apparent movement, of a jellyfish.
 

            The name “Cyro” is a combination of the first two letters of the species on which this robot’s design is based, Cyanea capillata, plus the first two letters of the word “robot.”   Cy + ro = Cyro.

            In 2012, the first prototype, RoboJelly, was developed.  About the size of man’s hand, this smaller version had about the same dimensions as an actual jellyfish.  But why model an Unmanned Underwater Vehicle, UUV, after a Jellyfish? 

            Biomimicry.

            Biomimicry describes any technology imitating (copied from) nature.  In other words, if you want a drone that works in a particular way, and the best example of such performance is a biological organism, imitation is the shortest distance to the goal.

            And the selection of a jellyfish as a model for this UUV was no accident.  The actual jellyfish, as an organism, seems “designed” to operate with a low metabolic rate allowing it function with remarkably low energy consumption.  Although jellyfish all share certain common traits of physical structure, these creatures come in many different shapes, sizes and colors.  This variety provides a wide selection of “models” for possible design imitation.

            But the desirable qualities of the jellyfish, for technological imitation, go farther than its fuel economy and assortment of “body styles.”  Jellyfish successfully live and function in all the major oceans of the world.  They flourish in warm tropical waters and as well as the colder waters of the arctic.  We’re most familiar with the coastal species, which are, in fact, the most numerous.  But a number of jellyfish species live (and live well) at depths of over four miles beneath the surface of the ocean.

            Virginia Tech Professor of Mechanical Engineering Shashank Priya heads the robotic jellyfish project.  Development is focusing on power consumption with the goal of extending Cyro’s operating time from hours to months. Alex Villanueva, a doctoral student in mechanical engineering, working under Priya, explained that the larger payload capacity will also allow more operating time and longer range. 

            Cyro is Virginia Tech’s contribution under a $5,000,000 grant from the U.S. Naval Undersea Warfare Center of the Office of Naval Research.  The grant is shared with UCLA, Stanford University, Providence College, and the University of Texas. The Navy’s ultimate goal is to develop “self-powering, autonomous machines,” which will operate in the oceans for purposes of surveillance, monitoring ocean currents, mapping the ocean floors and studying undersea life.

Thursday 15 May 2014

GCLM5444HOxenia

[Author’s Note: (trivia) Cyro is patterned after a jellyfish species commonly called the “lions mane.”  This extremely large jellyfish was featured in an original Sherlock Holmes short story, The Adventure of the Lion’s Mane.  But the story exaggerates the toxicity of this species’ sting, which is no more dangerous than the sting of the jellyfish often encountered in coastal waters.]

Virginia Tech researchers unveil large robotic jellyfish that one day could patrol oceans

Robotic jellyfish named Cyro could one day work for the Navy

Robotic Jellyfish ‘Cyro’ Could Work for Navy, Come After You

ROBOTS: Hydra – The Undersea Drone with Drones of its Own

1 May 2014

 

            Cut one head off and two more grow back?  It’s from Greek mythology.  The creature’s name was Hydra.  DARPA has given the same name to a planned unmanned vehicle.  This underwater drone would do little in terms of actual “engagement.”  Instead, it would be “stocked” with drones of every imaginable kind.  Traveling to various hot spots, the Hydra would deploy the numbers and kinds of drones needed to do the job – whatever that job might be.

            The Hydra has been called an underwater version of an aircraft carrier.  Not designed so much to engage in combat or reconnaissance, the Hydra, like the aircraft carrier, is intended to transport items designed for these very purposes.  But, unlike the aircraft carrier, the Hydra is both unmanned and a “submarine” vehicle – tagged with the acronym, UUV, unmanned underwater vehicle.

See: Photos of the Day: The Hydra underwater drone

            A DARPA sponsored presentation of the project was made on “Proposer’s Day” at John’s Hopkins University Applied Physics Laboratory.   The Hydra program’s goal was described as the development of an “unmanned air and undersea system” to deliver “unmanned air and underwater vehicles into operational environments.”

            When you think about it, you realize that building the Hydra also requires building a fleet of drones of every variety with extremely diverse functions.  Just to fill-in a few details, the Hydra doesn’t just carry its payload as it travels underwater, its drones are actually deployed underwater.  Some of the deployed drones could perform their functions underwater.  Others could rise to the surface and continue to operate as unmanned surface vehicles.  Still others could rise to the surface and take off into the air -- becoming airborne.  [view image]

            Although the Hydra is unmanned, this doesn’t mean that it couldn’t be equipped with deployable drones that could, themselves, transport human beings in emergencies. DARPA engineers are considering the design of a “submersible” “capsule” for the transportation of troops.  Again, the troops wouldn’t be passengers in the Hydra.  Rather, the Hydra would deploy a drone able to pick up troops at one location and take them to another location.  The resulting stealth delivery would assist in rescue operations as well as the surgical strike type of military operation. 

            The Hydra’s program manager, Scott Littlefield, has pointed out the economic savings resulting from the use of unmanned technology in the development of underwater defense strategies.  Littlefield sees these unmanned technologies as providing a way to expand our defense capabilities even with tightening budgets. 

DARPA goes deep: New Hydra project to see underwater drones deploying drones

            Clearly, drone technology would be much cheaper to deploy and operate than similar manned technologies.  Historically, submarines, for example, costly, inflexible (slow) in response compared to this proposed drone technology.  Submarines have also presented unavoidable dangers to human crews as well as a good deal of discomfort.

Hydra Program [view video]

            The sheer extent of underwater coverage promised by the development of the Hydra is unprecedented.  The hydra program would maintain a presence beneath all the relatively shallow waters of not only the seas, but also the world’s river systems. But what about the deep seas . . . ?

            Not to worry.

            Complementing the UVV’s of the Hydra program would be the “pods” and “modules” of the UFP program.  The UFP (upward falling payloads) program would place pods containing supply modules and drones on the deep sea floors.  Because of the extremely depth (more than two and a half miles), the pods would quite difficult to access.  On the one hand, the depth provides the perfect stealth.  On the other hand, the access difficulties make it necessary that the pods be designed and stocked to last years at a time. 

UFP [view image]

            Like the Hydra, the UFP pods would be filled with supply modules and drones that could be delivered to the ocean’s surface when needed.  And the delivery is the easiest part of the system.  The modules and drones would be buoyant, lighter than water.  Upon release, with water pressure at the deep sea floor, the drones and/or modules don’t just float, but seem to rush to the surface so quickly that it looks like . . . “falling upward.” Hence, “upward falling payloads.”

ROBOTS: Upward Falling? A New Generation of Undersea Drones

            The Hydra and UFP programs are part of a new wave of logistically oriented drone systems.  The new emphasis can be seen with another DARPA program to compliment the familiar aerial combat drone. DARPA, together with Lockheed Martin, is working to develop not only unmanned aircraft, but unmanned land vehicles to supply soldiers in combat. 

Thursday 1 May 2014

GCLM5444HOxenia

DARPA: Hydra Program

DARPA Hydra, here comes the mothership designed to deploy drones

What SNL did for the land shark, Lockheed Martin may be doing for the “land drone.”  See: Army, Lockheed to test drones-only mission, by air and land

 

ROBOTS: Upward Falling? A New Generation of Undersea Drones

1 May 2014

If you want to keep an eye on everything, the ocean is a problem.  How do you “watch it?”  It’s really big.  It’s water.  So, it’s hard to put things in it and make them stand still.  Also, watching the ocean is kind of boring.  Over most of the ocean surface, most of the time, nothing much happens.  Then, when you do find out something is “going on” in a particular place, you can go there, but you can’t get there fast enough.

Most of us would just ignore the whole thing and busy ourselves with something else.  But “problems” like the ocean (how to watch all of it all the time) are the meat and potatoes of DARPA strategists.

First, the problem of getting things to say put.  How can you avoid the “ocean issues?” (It’s liquid and wavy.)  If only you could find a solid surface.  But wait.  You can!  The bottom of the ocean.  Go down far enough, and you’ll hit the sea floor.  Then, you need to put something there.  What?   Something that will stay put.  But wait.  That “something” has got to able to “respond” quickly when needed.  So, if you tie it down to the sea floor, it won’t be able to move in an emergency.  If you make it so heavy that the ocean currents won’t be able to move it, it will take more power than you can supply when it has to move in an emergency.

Solution?

Nail down a “pod” to the sea floor.  Like the natural pod filled with seeds, this pod will be filled with minidrones.  When the need arises, release the drones.  Where will the drones get the power to move to the surface quickly?   Well, no “power” is really necessary.  You just make them buoyant – lighter than water.  So, instead of powering to the surface, they rise automatically – “falling upward.”

 

The pods are designed to rest on the ocean floor for long, long periods of time waiting to release drones that will not only rise “toward” the surface, working as sea drones, but will also rise to surface and take off into the air – as airborne drones.

Then, what are these drones supposed to do?

The “upward falling” drones would offer “non-lethal assistance.”  That is, these drones would have surveillance capabilities (surveillance sensors) providing intelligence or targeting information.  They could, also, act as decoys and even use their “low-power” lasers to attack.

The “low power” of the lasers and “non-lethal” attack capabilities are significant and intended limitations.  With over 50% of the ocean floor deeper than two and half miles, recovering the pods, once deployed would be difficult.  If the pods contained advanced weaponry or extremely hazardous materials, their dysfunction or deterioration could cause unintended and unwanted damage to ocean-going vessels.

Following DARPA’s guidelines, the success of the "Upward Falling Payloads" (UFP) program requires the development of a system that can do three things:  First, the system must be able to withstand the extreme pressure of the deep sea floors for a period of years.  Second, the system must be reliably triggered by remote control (“standoff command”).   And third, the drones must “fall upward” fast – rise through the water and deliver their payload.

UFP’s first phase began in 2013 with the design of the pods and their deployable drones/capsules.  Also, the design required communication capabilities allowing the pods to communicate among themselves.  DARPA is now taking bids for the final two phases of the Upward Falling Payloads (UFP) program.

The second phase includes the testing and demonstration of the developed prototypes at sea.  In the third and last phase, to be completed by early 2017, the pods and drones will be scattered at full depth and required to work as one system.  The actual testing will probably be done on either side of the Pacific Ocean with some testing in the western Pacific and other tests in the eastern Pacific off the U.S. coast.

Note the emphasis on communication among the pods and their payloads as well as the systematic performance of all the devices as a coordinated group.   Each unit is not intended or designed to operate with complete independence.  Rather, each is part of a system and network of deep sea pods with the ability to communicate with each other to allow multiple pods to coordinate their activities – if and when necessary.

These units have a natural camouflage/stealth.  Their depth will make detection difficult.  In fact, the pods, themselves, will serve out their useful “lives” well below the depth at which manned vehicles can operate.  So no one can or will be dropping by and visiting the pods on a regular basis.

Someday, maybe sooner than we think, the deep ocean floors will be covered with a latticework of pods quietly biding their time until they are needed.

DARPA UFP’s Upward Falling Payloads

DARPA Wants to Seed the Ocean Depths With Upward Falling UAV Pods

Thursday 1 May 2014

GCLM5444HOxenia