University students design ice cream serving robot

A robot that can automatically serve ice cream with selected toppings and pass on a spoon? 26 students at Ohio Northern University under supervision of their professors have built an ice cream serving robot consisting of two Kuka industrial robots which man their stations inside a kiosk of custom-fabricated components. Check out the video after the break, what they have achieved in this project is brilliant!

UNNATURAL ROBOTS

LIVING creatures took millions of years to evolve from amphibians to four-legged mammals - with larger, more complex brains to match. Now an evolving robot has performed a similar trick in hours, thanks to a software "brain" that automatically grows in size and complexity as its physical body develops.

Existing robots cannot usually cope with physical changes - the addition of a sensor or new type of limb, say - without a complete redesign of their control software, which can be time-consuming and expensive.

So artificial intelligence engineer Christopher MacLeod and his colleagues at the Robert Gordon University in Aberdeen, UK, created a robot that adapts to such changes by mimicking biological evolution. "If we want to make really complex humanoid robots with ever more sensors and more complex behaviours, it is critical that they are able to grow in complexity over time - just like biological creatures did," he says.

As animals evolved, additions of small groups of neurons on top of existing neural structures are thought to have allowed their brain complexity to increase steadily, he says, keeping pace with the development of new limbs and senses. In the same way, Macleod's robot's brain assigns new clusters of "neurons" to adapt to new additions to its body.

The robot is controlled by a neural network - software that mimics the brain's learning process. This comprises a set of interconnected processing nodes which can be trained to produce desired actions. For example, if the goal is to remain balanced and the robot receives inputs from sensors that it is tipping over, it will move its limbs in an attempt to right itself. Such actions are shaped by adjusting the importance, or weighting, of the input signals to each node. Certain combinations of these sensor inputs cause the node to fire a signal - to drive a motor, for example. If this action works, the combination is kept. If it fails, and the robot falls over, the robot will make adjustments and try something different next time.

Finding the best combinations is not easy - so roboticists often use an evolutionary algorithm to "evolve" the optimal control system. The EA randomly creates large numbers of control "genomes" for the robot. These behaviour patterns are tested in training sessions, and the most successful genomes are "bred" together to create still better versions - until the best control system is arrived at.

MacLeod's team took this idea a step further, however, and developed an incremental evolutionary algorithm (IEA) capable of adding new parts to its robot brain over time.

The team started with a simple robot the size of a paperback book, with two rotatable pegs for legs that could be turned by motors through 180 degrees. They then gave the robot's six-neuron control system its primary command - to travel as far as possible in 1000 seconds. The software then set to work evolving the fastest form of locomotion to fulfil this task.

"It fell over mostly, in a puppyish kind of way," says MacLeod. "But then it started moving forward and not falling over straight away - and then it got better and better until it could eventually hop along the bench like a mudskipper."

When the IEA realises that its evolutions are no longer improving the robot's speed it freezes the neural network it has evolved, denying it the ability to evolve further. That network knows how to work the peg legs - and it will continue to do so.

At this point, it is just like any other evolved robot: it would be unable to cope with the addition of knee-like joints, say, or more legs. But unlike conventional EAs, the IEA is sensitive to a sudden inability to live up to its primary command. So when the team fixed jointed legs to their robot's pegs, the software "realises" that it has to learn how to walk all over again. To do this, it automatically assigns itself fresh neurons to learn how to control its new legs.
When the team fixed jointed legs onto the robot, it 'realised' it had to learn how to walk all over again 

As the IEA runs again, the leg below the "knee" is initially wobbly, but the existing peg-leg "hip" is already trained. "So it flops about, but with more purpose to it," says MacLeod. "Eventually the knee joint works and the robot evolves a salamander-like motion."

A Robomedic for the Battlefield

A snakelike robotic arm may one day medically attend to soldiers as they are carried off the battlefield.The first 30 minutes after a battlefield injury are dire: that's when nearly 86 percent of battlefield deaths occur. Before attending to the wounded, frontline physicians have to quickly locate the casualty and extract him from the battlefield, often under heavy fire. This can take up costly minutes, as well as expose medics themselves as possible targets.

Now researchers at Carnegie Mellon University (CMU) are developing technology to give battlefield medics a helping hand--literally. Howie Choset, an associate professor of robotics at CMU, has engineered a snakelike robotic arm equipped with various sensors that can monitor a soldier's condition. The robot can be wirelessly controlled via a joystick, so that a doctor at a remote clinic may move the robot to any point on a soldier's body to assess his injuries as he's being carried to a safe location. The robot's serpentine flexibility allows it to maneuver within tight confines, so that, in case a casualty can't be extracted from the battlefield immediately, the robot can perform an initial medical assessment in the field.

Choset and his colleagues have been building "snakebots" for over 10 years, improving range of motion and flexibility, as well as minimizing the overall size in multiple prototypes. In the past, the group has designed robots for urban search-and-rescue missions, and has worked with Ford Motor Company to build snake robots for precise auto-body painting. The team recently formed a startup company to commercialize one of its latest technologies, a robot that can potentially perform heart surgery.

Currently, the team is collaborating with the U.S. Army's Telemedicine and Advanced Technology Research Center (TATRC) to integrate the robotic arm within the military's high-tech stretcher, called the Life Support for Trauma and Transport system (LSTAT). This stretcher is essentially a portable intensive-care unit, with a ventilator, defibrillator, and other physiological monitors, and it's currently being used in areas of Iraq and Afghanistan. Medics can quickly load a casualty onto the stretcher and attend to injuries with the equipment onboard.

"It has all these sensors onboard so we can perform preliminary diagnostics and maybe therapeutics to save the guy's life," says Choset. "The problem is, these sensors are attached to the LSTAT, and you would have to move them by hand, and if someone's shot and you go over and help them, you're an easy target. So we want to automate this whole system, and robotically move the sensors onto the patient while he's being dragged off the battlefield."

ROBOTICS FOR PERSONAL ASSISTANCE,MEDICAL

Biomedical robots performed less invasive and more complex experimental surgeries, winged robots copied each other to perform potential military maneuvers, and researchers began work on robots that may even be able to travel through the blood to zap a tumor. Some highlights:

Grab and Grasp:

Robotic grasping and learning is becoming sophisticated enough that people may soon be able to simply gesture to any object that they want and, without needing to program specifics, rely on a robot to retrieve it. A robot demonstrated this year at Georgia Tech, El-E (pronounced "Ellie"), a wheeled, one-armed robot, follows a green laser pointer to retrieve objects.  Later in the year, the group gave El-E new abilities based on how dogs respond to humans.  Another grasper, the UMass Mobile Manipulator--UMan, for short--demonstrated that it could learn how to use new objects.  Just as humans learn by testing an object, UMan is able to experiment and learn by playing with objects, including scissors, shears, and wooden toys.

Stomach Explorers:

While doctors have used capsule cameras for the past few years to image the insides of patients, they hope for ways to control such a camera so that it pauses at areas of interest. A group in Germany uses a magnetic device outside the body to control the movement of a pill camera  while researchers at Carnegie Mellon University created a robot capsule that can anchor on delicate internal tissue without damaging it.