The Japanese Ministry of Defense recently unveiled this new flying surveillance drone, which it says will be deployed to disaster areas such as volcanic eruptions and landslides.

While my Japanese is shaky (and by that, I mean non-existent), judging from the translation of the caption at YouTube, it looks like operators can program this robot with a pre-determined altitude and route via GPS where it can operate for up to 30 minutes, transmitting images back to a computer all the while.

(Via Crave)

iRobot, a company perhaps best known for their line of Roomba robotic vacuum cleaners, also produces several military robots that are used for activities ranging from simple surveillance to bomb disposal.

The biggest and baddest military bot, the 710 Warrior, is not only rugged and strong (capable of carrying up to 150 pounds), but also shows a remarkable degree of mobility.

In the above video, iRobot crew is able to easily coax the machine up a steep 60-degree flight of stairs, down what looks to be about a one-foot drop, and then back up. It looks like the operator’s ability to cleverly shift its payload and change its center of balance is crucial to the Warrior’s ability to get around.

(Via @BotJunkie)

Yesterday defense contractor Raytheon released the above video for the latest iteration of their XOS 2 exoskeleton, which coincides with the home release of Iron Man 2. The XOS 2 is designed to help military personnel personnel by lightening their loads and assisting with lifting heavy objects, like armaments.

A Raytheon spokesman said he expects to deploy the non-tethered XOS 2, which is powered by an internal combustion engine, in the field in the next “3 – 5 years.”

In the meantime, let’s get working on the rocket boots and repulsor blasters so we can get them online in time for the unveiling of the XOS 3, hmm?

(Via Engadget)

This video of humanoid robot iCub learning to accurately shoot a bow and arrow is interesting – the robot is evaluating each missed shot, and then making adjustments to improve its aim. It takes it eight tries to finally hit the target.

The learning algorithm, called ARCHER (Augmented Reward Chained Regression) algorithm, was developed and optimized specifically for problems like the archery training, which have a smooth solution space and prior knowledge about the goal to be achieved. In the case of archery, we know that hitting the center corresponds to the maximum reward we can get. Using this prior information about the task, we can view the position of the arrow’s tip as an augmented reward. ARCHER uses a chained local regression process that iteratively estimates new policy parameters which have a greater probability of leading to the achievement of the goal of the task, based on the experience so far. An advantage of ARCHER over other learning algorithms is that it makes use of richer feedback information about the result of a rollout.

While we’re talking robot archery, here’s a neat little video of toy humanoid robot i-SOBOT shooting a mini bow and arrow. There’s no calculation going on here to figure out where it should aim, but hey – robots with tiny weapons are still cool.

(iCub story via BotJunkie)

HULC, or Human Universal Load Carrier, is an exoskeleton designed by Lockheed Martin to assist soldiers carrying heavy loads in the field. According to its manufacturer, HULC allows soldiers to carry up to 200 pounds for extended periods of time while preserving the user’s range of movement – including “deep squats, crawls and upper body lifting.”

As you can see in Lockheed’s promotional video above, HULC isn’t a bulky exoskeleton that provides strength in exchange for mobility. Users appear to be able to move quickly and easily even over rough terrain.

The U.S. Army is apparently intrigued by the concept, and this week announced a $1.1 million contract to actually field a few HULC units among active troops. While $1.1 million is a relative drop in the bucket as far as military spending goes, it may foretell a future in which assistive exoskeletons become common in the military for logistics staff and even soldiers engaged in combat.

New Zealand firm Rex Bionics took a different approach with their “robotic legs,” which are designed to allow the disabled to walk. The company notes the Rex is not intended to be a wheelchair replacement, but is instead a way to augment a chair – for instance, when needing to travel up and down stairs or access items stored at a standing height.

Having been in development for the last seven years. the Rex is expected to go on sale “soon” in its home country and is expected to cost $150,000.

Today the British Ministry of Defense unveiled a new bomb disposal robot called a “Dragon Runner” that breaks new ground in portability – the system is a mere 14 pounds and small enough to fit in a backpack. Despite its small size, however, it sounds remarkably capable:

The sophisticated robot is manoeuvred over tough terrain using an intuitive controller, not unlike to a games console.  It can climb stairs and even open doors.

The dragon has four on board cameras that relay images back to the operator via the hand held controller. Each shot can be viewed on screen separately or all four at once in a grid.

It is fitted with a manipulator arm that can dig around suspicious objects and lift items that weigh up to 10lb. It can also place small charges that work to disrupt suspect devices.

Manufacturer QuinetiQ has built about 100 of the robots to date, which will be used by a “number” of British regiments.

(Via Instapundit)

This week DARPA announced its “Mind’s Eye Program,” an effort to research visual intelligence in machines. More specifically, it will serve to not only enable cameras to recognize objects (an ability some machines are already capable of) but determine interactions between objects it sees, and then use this information to provide a “complete narrative of action.” Naturally, the military applications of this technology would be profound (PDF):

One of the desired military capabilities resulting from this new form of visual intelligence is a smart camera, with sufficient visual intelligence that it can report on activity in an area of observation. A camera with this kind of visual intelligence could be employed as a payload on a broad range of persistent stare surveillance platforms, from fixed surveillance systems, which would conceivably benefit from abundant computing power, to camera‐equipped perch‐and‐stare micro air vehicles, which would impose extreme limitations on payload size and available computing power.

DARPA will be conducting an “Industry Day” in April to familiarize industry with the idea and explain program requirements.

(Via Wired)

There are a total of seven challenges on which they are calling for solutions:

1. Plan and execute approach and regress routes within both urban and wilderness terrain, and without preloaded maps or terrain models
2. Communicate with and facilitate communications between patients and human medics
3. Execute command, control, and coordination of individual robots, robot teams, and medical payloads
4. Perform remote/stand-off initial casualty assessment to iidentify injuries sufficient to prevent further injury during robotic casualty extraction
5. Lift, move, drag, tow, or otherwise effect recovery of patients in any of numerous body positions from hazardous environments in any terrain to safe locations
6. Provide closed loop or semi-autonomous casualty monitoring and enroute care sufficient to mitigate risk associated with “abandonment” concerns during unattended CASEVAC (casualty extraction and evacuation – ed.)
7. Plan and conduct recovery from errors or the unexpected

The posting also notes that the Army has been working on research in this area, but is having particular trouble solving the issue of safely picking up wounded soldiers for transport (number five). From the information in the posting, as well as the added questions and answers, any proposals that could come up with a solution to this particular issue would have the best chance at a positive evaluation.

(Via New Scientist)

BBTV3NNBR7KQ

While most news sources note that the drones will be used for “surveillance,” they also mention the craft can carry “diverse payloads.” In the news stories I’ve read, Israeli military officials haven’t commented extensively on the platform’s bombing capabilities, but have noted that it is “versatile.”

Given its size and capabilities, I wonder if Eitan will provide Israel with the long-range bombing capacity the IAF has lacked – until now.

When fighting in rough, difficult-to-traverse terrain (such as that found across much of Afghanistan), the military relies on helicopters to quickly deliver weapons and ammunition to soldiers on the battlefield. Unfortunately, manned helicopters are expensive and make attractive targets for enemy fire.

Enter the K-MAX Unmanned Multi-Mission Helicopter, a drone that passed a test devised by the U.S. Marines that included (PDF warning):

… hovering at 12,000 ft. with a 1,500-pound sling load; delivering 3,000 pounds of cargo well within the six-hour required timeframe to a forward operating base (two 150 nm round-trip flights); remotely controlling flight and a precision load delivery by a ground-based operator in both day and night conditions; and uploading a new mission plan to the aircraft’s mission management system during flight.

The most impressive part is not necessarily that the chopper was merely unmanned, as the U.S. has been effectively using unmanned drones in war for years. Rather, the chopper passed the test nearly autonomously, successfully delivering cargo to a series of pre-programmed locations. It was only tested through manual ground control for one delivery at the Marines’ request.

In total, the demonstration cost $860,000, which apparently is a downright bargain as far as military operations go. For the sake of troops on the ground, the faster they get these birds deployed in the field, the better.