Tell us what apps you want.
In July of 2008, the app store concept was launched by Apple and it has dramatically changed the mobile industry. Over 1.2 billion apps were downloaded between Christmas and New Year according to Flurry Analytics. That’s about 60 percent higher than every week of December before Christmas, and marks the largest number of apps ever downloaded in a single week.

Robot app stores are now beginning to happen as well and visions of apps for tele-robotic safety and surveillance pop into mind, or light domestic duties, or…? My imagination runs wild at the thought and I imagine these new apps being easy to use, having multiple functions and a low cost.

But that is then and now is now. At present there are three significantly different approaches to an app store for robots (although additional “stores” are in the planning stage and many existing stores are making arrangements to expand beyond their own boundaries (like iRobot’s online store and Willow Garage’s application chooser)):

1. MyRobots, by Canadian RobotShop, has opened a combination app store and online environment to permit robotics products to be monitored and updated seamlessly as well as providing a Cloud repository for applications that run from the Cloud instead of the robot itself. MyRobots’ slogan is “Connect your robots; reap the benefits.” The online monitoring aspect of MyRobots is similar to the MyMotoman Remote Monitoring Service initiated my Yaskawa Motoman America in August.
2. Robots App Store, by San Francisco serial entrepreneur Elad Inbar, is developing their store to be an online marketplace for selling apps and also providing developers their own sales portal. Robots App Store, in addition to attempting to trademark the hyphenated words “Robot-app”, has a wonderfully informative info-graphic explaining the need and market for robot apps. What it doesn’t explain is how the process will be different than what most of us have come to expect when buying apps from Apple’s App Store. It involves tinkering with development tools and operating system platforms – a turnoff for all but the most avid hobbyist.
3. RoboEarth, an EU public-private consortium, is using the Internet to create a giant open source network and database repository that can be accessed and updated by robots around the world. Focused more on academia and industrial applications, RoboEarth speeds up robot learning and adaption in complex tasks and can execute tasks that were not explicitly planned for when the robot was designed.

Service robotics – as differentiated from industrial or defense/security/space robotics – needs to provide inexpensive and efficient assistance to humans to be profitable in the consumer marketplace and emerge as a serious industry. We already have vacuum robots and adaptive cruise control bots, but there are thousands of other possible applications. Entertainment, embedded systems, eldercare and home assistance are certain to be amongst the first groups to benefit as the service robotics industry materializes.

There are two different approaches to providing those services:
C3P0
The Ultimate Humanoid Robot

1. A universal robotic tool that will be able to perform all the tasks we ask it to do, including social communication – a robot like C3P0. This is the goal of humanoid robots. At this time, humanoid robots are costly, difficult to control, slow and intrusive. Although human-robot communication is part of the lure, and there is much progress in that area, we are at least five years away from devices that not only speak but understand. IBM’s Watson and Apple’s Siri are proving that it can be done but their software isn’t yet available to front-end robotic applications and there is, at present, little comparable in the robotics world (although the recent ABC Nightline interview of Hanson Robotics’ Philip Dick robot shows there is serious ongoing development). Even when the two-way communication aspect is further developed and made available to roboticists, it will be many more years until the goal of a C3P0-like sophisticated humanoid robot will be ready.

2. A more near-term and progressive solution is to forego universality and design specific devices that provide limited, task-specific means to communicate and complete specific tasks… think Roomba.

Certainly #2 has more near-term promise and is where most of the market is focused for the rest of this decade. As service robots begin to meet the four core criteria of low cost, low intrusiveness, safety and modularity – which is happening almost daily with technology changes – an incredible number of opportunities for service robot apps can be foreseen. The key will be to make them practical, affordable and acceptable by humans as part of their day-to-day life.

Robots App Store:

At discussions with Elad Inbar, CEO of Robots App Store, I asked about the size of the marketplace and was surprised to find that there are 14 million service robots out there now, and it’s estimatited that by the end of 2014 there will be 24 million. If only 15% of them were able to buy two apps at an average price of $9.99, that’s $72 million in revenue. If they each bought five apps instead of two, the figure would be $180 million. If 20% purchased five apps it would be a $¼ billion business. How real and applicable those figures are depends on many complex variables: willingness of the buyer to tinker, quality of the apps, flexibility of the robots, price, disposable income, etc. In fact, apps for robots with an Apple or Android front end will not be available in the Robots App Store but will onlybe available through the Apple (or Android) App Store – which limits the market metrics… an example of this Apple exclusivity is the hundreds of thousands of Parrot’s AR.Drone quadcopters.

Regardless of the metrics, the process of developing and sharing apps can provide a giant leap toward rejuvenating the robots we have and teaching the audience (the buyers) that they can get what they want from their robot(s). Thus the buyer becomes more educated and demanding and the producers less capricious — and more profitable.

From an infographic by Robots App Store showing possible app prices for various robots
Inbar’s Robots App Store is “focused solely on the personal and service robots marketplace because “industrial robots are very limited in their functionality… not to mention that they are oriented to quickly repeat the same tasks 24/7 unlike personal and service robots that can be extended to virtually everything! I believe in the consumers market – as they will grow like a hockey-stick!”

When I asked what kind of apps we will see once the store opens, Elad said, “We have more than 100 approved robotic developers working on apps for Romba, Pleo, AIBO, NAO, DARwin and others arranged in categories like Entertainment (dances, etc.), Story Telling (for kids), Utilities (like home search engines), Tools (like face recognition and such) and many others.”

Finally, I asked about the mechanics of porting an app or game from one robot to another. “ROS by Willow Garage, Gostai, URBI and others are trying to build a standard open platform OS that will run across all robots. We, as a marketplace, don’t take sides in this battle. We are serving every robot, ever developer, and every OS. But surely we’ll have to take sides one day when the critical mass will decide which is the widely spread and used OS…. Personally I believe that ROS is in the best position to become the de-facto standard and I personally love to work with their code. It is easy, simple and their libraries can be ported quickly between different robots – a huge advantage for every robot apps developer.”

MyRobots:

The goal for MyRobots is to get all robotic platforms to collaborate and operate together (regardless of their operating system) via the use of Cloud robotics. Although like other stores MyRobots will sell downloadable apps too, their main interest is to develop applications that will run on the Cloud thereby freeing computation time and resources on the robot and will make a single app compatible with many different robotic platforms.

Says MyRobot’s Mario Tremblay:

Cloud Robotics is the application of the cloud computing concept to robots. This means using the Internet to augment the robots capabilities by off-loading computation and providing services on demand. Being connected to the cloud also helps robots to collaborate with other machines, smart objects and humans. Through this collaboration, robots transcend their physical limitations and become more useful and capable since they can delegate parts of their tasks to more suitable parties. By combining increased communication capabilities to the ability and flexibility of running and storing part of their intelligence (i.e. software, behaviors and apps) on the MyRobots infrastructure, smart objects and robots become augmented and constitute a new and revolutionary concept for the future of robotic intelligence: Cloud Robotics.

RoboEarth:

RoboEarth is an EU-funded consortium of educators, research facilities and corporations working together to provide a world-wide-web for robot applications. RoboEarth, in development for the last two years, can be seen as a giant network and database repository where robots share information and learn from each other about their behavior and environment thereby showing a better way of robot collaboration and information sharing. RoboEarth’s premise is to out-source application development by tapping creativity and experience from wherever they can. Further, they believe (as do I) that more apps sold equates to more robots sold. Quoting Markus Waibel:

The central beauty of the App Store idea is that it enables a feedback cycle: The more devices (phones/robots) you sell, the more lucrative the app store becomes for 3rd party developers to program new apps. And the more new apps, the more devices you sell. In my opinion, robotics does not yet have critical mass to jump start this at the present time.

RoboEarth is fast becoming a large shared repository for maps, object models, task descriptions, robot models, semantic relations, probabilistic relations and “action recipes” (apps) within the global academic community. It remains to be seen whether the big robot manufacturers will embrace the technology for their products.

One unintended consequence of RoboEarth’s efforts may be within the used robots marketplace. Imagine the financial gain if an older, unused robot is rejiggered to have new equipment as well as easy to train software to put the robot back to work on new projects. Another possible beneficiary is the end user who will be less likely to retain an intergrator/consulting service if the robot can handle ongoing changes on its own.

The ultimate goal of RoboEarth is to enable robots to share and store what they discover about the world on their own Internet.
**********
As a personal aside, I live in a rural area outside of a medium-sized city. 4G LTE is planned for 2012-2013 for the city but not my area and I have to pay for a very unreliable T1 line to my office. It’s hard to imagine having a service robot doing its thing in my world while relying on the Cloud (as represented by my T1). Maybe the concept will grow on me with time but right now I’m wary and a bit afraid of the consequences.
**********
What do you want from a robot app store? Let’s compile a long list of “wants” and send them off to these app stores and see whether they can get them for us. Why should we wait to see if our app magically appears in their store? Let’s ask for what we want. Send them to me and I’ll combine them all and send them off to the stores.

Here’s what I would like:

* For my old Pleo, anything that makes it do things it doesn’t already do.
* For my AR.Drone, a programmable camera shoot over a grid that I draw onscreen from a starting point where I place the drone.
* For my little iSobot, a complicated routine of fetch, grab and return.
* And for my new Swivl, something sexy where it follows my girlfriend around the room and refocuses as the distance changes.

But I also want to buy a used two-armed Yaskawa Motoman and put it to work in my workshop doing ad hoc tasks. I’d love to be able to add a hardware component add-on (or two), be able to use ROS or some simulation software to test my apps, buy some additional apps that do the kind of tasks I need to do in the shop, and have all that work seamlessly and without line-for-line coding, excessive cost or too much fiddling.

What do you want? Send them to me and I’ll combine them all and send them off to the stores.
Posted by Frank Tobe

From the reviews:

“This simply titled work is a result of carefully delivered lectures … for multiple classes of undergraduate engineering students over more than a decade. This exemplary course resource can serve as a basis for the study of robotics. Pedagogically well structured, it covers the fundamentals of industrial robotics … . this book is a concise, readable reference source for those getting their feet wet in the field. … Summing Up: Highly recommended. Lower- and upper-division undergraduates, two-year technical program students, and general readers.” (G. Trajkovski)

http://www.springer.com/engineering/robotics/book/978-90-481-3775-6

On Nov 8, 2011 Honda unveiled an all-new ASIMO humanoid robot newly equipped with the world’s first autonomous behavior control technology. With a further advance in autonomy, the all-new ASIMO can now continue moving without being controlled by an operator. ASIMO took another step closer to practical use in an office or a public space where many people come and go.

World’s first robotics application marketplace empowers developers with a platform to sell their apps, and provides consumers an efficient and secure method to purchase them.

Personal and service robots are on the rise. In fact, the number of robot owners is doubling each year, and units owned are expected to reach 100 million by 2017.  The soon to be $17 billion personal robotics industry is ripe with opportunity and creativity, which is why RobotsAppStore.com has created the first marketplace that connects robotics developers and consumers.

The online portal, which was filed two patents for its repository and security methods, sees remarkable parallels to the mobile apps industry and forecasts similar growth. After only three years, mobile app stores have increased from 500 to nearly a million apps available for purchase, with $3 billion paid to developers.

By providing the connecting medium that synergizes the robotics software industry, RobotsAppStore.com is hoping to launch personal robotics into a new era of creativity and sophistication.

In the store, you can find the most creative robotic apps on the planet, including laundry folding, pet feeding and soccer playing applications for over 50 kinds of robots, ranging from vacuum cleaners and dogs to humanoids and vehicles.

Currently, RobotsAppStore.com is seeking additional developers around the globe to submit their apps, so that they can become available for safe and efficient consumer purchase. The website is scheduled for consumer launch in late December or early January.

Assuring the safety of intellectual property by blocking theft of apps is central to RobotsAppStore.com’s mission. The company, which is headed by robot enthusiast and expert Elad Inbar, open the doors to its new San Francisco office this month.

LAW 0:

A robot may not injure humanity or, through inaction,allow humanity to come to harm (added later).

LAW 1:

A robot may not injure a human being, or, throughinaction, allow a human being to come to harm.

LAW 2:

A robot must obey the orders given it by human beingsexcept where such orders would conflict with the First Law.

LAW 3:

A robot must protect its own existence as long as suchprotection does not conflict with the First or Second Law.

Built by Vietnamese firm TOSY this beast towers at 1.88m in height, 120kg in weight with 39 degrees of freedom is a bi-ped with fast reactions a low inertia mechanical system and probably has a mean poker-face too.

Robotics is a huge field, from industrial robots to home robots to medical robots and more. Dan Barry’s favorite space in this field is what are called autonomous robots. These are robots that aren’t a single task machine, but creations that perform many functions, adapt to their environment and can work independently. We have the crude beginnings of autonomous robots now, but to move forward, Artificial intelligence must advance. Artificial Intelligence, or A.I., is the software that programs the robot. His brain.

Progress in AI, along with the rapid reduction in the prices of robotic components, will bring us robot companions sooner than most expect. Even if the initial versions are not very “human-like” they will still be able to do such things as go for a walk with you, help you if you’re hurt, teach you things, and soon enough will even get you that beer. He is so confident of the companionship aspect that he says that you will fall in love with your robot.

Will robots ever become self-aware, conscious? Barry doesn’t know for sure, but hopes that consciousness is an emergent property. If emergent, it’s only a matter of time.

“The coverage is unparalleled in both depth and breadth. No other text that I have seen offers a better complete overview of modern robotic manipulation and robot control.” — Bradley Bishop, United States Naval Academy Based on the highly successful classic, Robot Dynamics and Control, by Spong and Vidyasagar (Wiley, 1989), Robot Modeling and Control offers a thoroughly up-to-date, self-contained introduction to the field. The text presents basic and advanced material in a style that is at once readable and mathematically rigorous. Key Features * A step-by-step computational approach helps you derive and compute the forward kinematics, inverse kinematics, and Jacobians for the most common robot designs. * Detailed coverage of vision and visual servo control enables you to program robots to manipulate objects sensed by cameras. * An entire chapter on dynamics prepares you to compute the dynamics of the most common manipulator designs. * The most common motion planning and trajectory generation algorithms are presented in an elementary style. * The comprehensive treatment of motion and force control includes both basic and advanced methods. * The text’s treatment of geometric nonlinear control is more readable than in more advanced texts. * Many worked examples and an extensive list of problems illustrate all aspects of the theory. About the authors Mark W. Spong is Donald Biggar Willett Professor of Engineering at the University of Illinois at Urbana-Champaign. Dr. Spong is the 2005 President of the IEEE Control Systems Society and past Editor-in-Chief of the IEEE Transactions on Control Systems Technology. Seth Hutchinson is currently a Professor at the University of Illinois in Urbana-Champaign, and a senior editor of the IEEE Transactions on Robotics and Automation. He has published extensively on the topics of robotics and computer vision. Mathukumalli Vidyasagar is currently Executive Vice President in charge of Advanced Technology at Tata Consultancy Services (TCS), India’s largest IT firm. Dr. Vidyasagar was formerly the director of the Centre for Artificial Intelligence and Robotics (CAIR), under Government of India’s Ministry of Defense.

Robot intelligence has become a major focus of intelligent robotics. Recent innovation in computational intelligence including fuzzy learning, neural networks, evolutionary computation and classical Artificial Intelligence provides sufficient theoretical and experimental foundations for enabling robots to undertake a variety of tasks with reasonable performance. This book reflects the recent advances in the field from an advanced knowledge processing perspective; there have been attempts to solve knowledge based information explosion constraints by integrating computational intelligence in the robotics context.

This book describes the development of a robot mapping and navigation system inspired by models of the neural mechanisms underlying spatial navigation in the rodent hippocampus. Computational models of animal navigation systems have traditionally had limited performance when implemented on robots. The aim of the work was to determine the extent to which hippocampal models can be used to provide a robot with functional mapping and navigation capabilities in real world environments. The focus of the research was on achieving practical robot performance, rather than maintaining biological plausibility.