Category Archives: Robotics

Pakistani six member team to participate in OIC Robotics Challenge

ISLAMABAD: Six-member Pakistani team has arrived in Tashkent, Uzbekistan for participating in the first `OIC Robotics Challenge (ORC) which is being held within the framework of INNOWEEK of the government of Uzbekistan from October 27 to 31.

From Pakistan, Ali Raza from National University of Science and Technology (NUST) and Usama Naseer from COMSATS University, Islamabad will participate in Robo Sumo competition while Muhammad Awais Khan, Muhammad Ahmad Ali, Abdul Waqas and Muhammad Ammar Hassan from Lahore Garrison University, Lahore will participate in the Robo Football competition.

The event is being arranged by COMSTECH in partnership with the Organization of Islamic Cooperation (OIC) General Secretariat, Ministry of Innovative Development of the Republic of Uzbekistan and the Islamic Development Bank (IDB).

The ORC takes place in the UZEXPOCENTRE exhibition center Tashkent and is restricted to participants from OIC Member States with ages between 17 and 23 years, an official of COMSTECH informed.

The teams, selected by OIC Jury, consisting of three members and each team can be accompanied by a senior person as Team Manager or Mentor who will need to support his own travel.

The participants would have the opportunity to have a guided tour of some of the exciting INNOWEEK activities being held concurrently, the official said.

The competition will be held in two different categories, and first three winning teams will be given the prize money in each category.

The total prize money for the competition is $21,000 which will be distributed among winners of the competition in two categories including Robo Sumo and Robo Football.

The first three winners of the Robo Sumo category will be given $ 5,000, $ 3,000 and $ 1,000 respectively while the first three winners of Robo Football category will be given $ 6,000, $ 4,000 and $ 2,000 respectively, the official added.

View original post here:

Pakistani six member team to participate in OIC Robotics Challenge - The News International

Combining the skill of LEGO building with the knowledge of coding, teams throughout Delaware are preparing for the 2020 FIRST LEGO League.

One new entry will be the combined forces of Great Oaks Charter School and Kuumba Academy in Wilmington, as they held their first practice Thursday afternoon.

"I've always loved technology and coding, so I came here from a poster. It said LEGO League, and I thought 'oh, okay', and it has coding, so it's great," saidNasir Briscoe, an eighth grade student at Great Oaks.

LEGO League is split into four different age groups covering K-12, and challenges teams to build a robot, and then pre-program its motor to do various tasks, such as moving objects, or in one case, helping get a person unstuck so they can continue swinging.

Briscoe said he can still remember when he got his first set.

"I didn't really play with toys as a kid, I wasn't one of those kids. I played with technology, and when she gave me a LEGO set, I started building it, and I loved it."

Article continues below advertisement

The Great Oaks/Kuumba team is getting a huge assist from Mount Pleasant junior Davis Miller, who is serving as the team's mentor, while also helping to get his school's team ready for the top division.

"What gets me into robotics is the fact probably half of it is team-building and the camaraderie of the students that you wouldn't get anywhere else, except maybe a sports team or a family. I think what it teaches kids about working together is invaluable."

Teams, including traditional schools along with various Boys & Girls Clubs in Delaware, are in the preliminary planning and building stages through the end of 2019. Statewide competition begins in January, and successful teams can move on to national, and perhaps even international, competition.

Read more:

LEGO-based robotics teams growing in Delaware | The Latest from WDEL News - WDEL 1150AM

MIT researchers have compiled a dataset that captures the detailed behavior of a robotic system physically pushing hundreds of different objects. Using the dataset the largest and most diverse of its kind researchers can train robots to learn pushing dynamics that are fundamental to many complex object-manipulation tasks, including reorienting and inspecting objects, and uncluttering scenes.

To capture the data, the researchers designed an automated system consisting of an industrial robotic arm with precise control, a 3D motion-tracking system, depth and traditional cameras, and software that stitches everything together. The arm pushes around modular objects that can be adjusted for weight, shape, and mass distribution. For each push, the system captures how those characteristics affect the robots push.

The dataset, called Omnipush, contains 250 different pushes of 250 objects, totaling roughly 62,500 unique pushes. Its already being used by researchers to, for instance, build models that help robots predict where objects will land when theyre pushed.

We need a lot of rich data to make sure our robots can learn, says Maria Bauza, a graduate student in the Department of Mechanical Engineering (MechE) and first author of a paper describing Omnipush thats being presented at the upcoming International Conference on Intelligent Robots and Systems. Here, were collecting data from a real robotic system, [and] the objects are varied enough to capture the richness of the pushing phenomena. This is important to help robots understand how pushing works, and to translate that information to other similar objects in the real world.

Joining Bauza on the paper are: Ferran Alet and Yen-Chen Lin, graduate students in the Computer Science and Artificial Intelligence Laboratory and the Department of Electrical Engineering and Computer Science (EECS); Tomas Lozano-Perez, the School of Engineering Professor of Teaching Excellence; Leslie P. Kaelbling, the Panasonic Professor of Computer Science and Engineering; Phillip Isola, an assistant professor in EECS; and Alberto Rodriguez, an associate professor in MechE.

Diversifying data

Why focus on pushing behavior? Modeling pushing dynamics that involve friction between objects and surfaces, Rodriguez explains, is critical in higher-level robotic tasks. Consider the visually and technically impressive robot that can play Jenga, which Rodriguez recently co-designed. The robot is performing a complex task, but the core of the mechanics driving that task is still that of pushing an object affected by, for instance, the friction between blocks, Rodriguez says.

Omnipush builds on a similar dataset built in the Manipulation and Mechanisms Laboratory (MCube) by Rodriguez, Bauza, and other researchers that captured pushing data on only 10 objects. After making the dataset public in 2016, they gathered feedback from researchers. One complaint was lack of object diversity: Robots trained on the dataset struggled to generalize information to new objects. There was also no video, which is important for computer vision, video prediction, and other tasks.

For their new dataset, the researchers leverage an industrial robotic arm with precision control of the velocity and position of a pusher, basically a vertical steel rod. As the arm pushes the objects, a Viconmotion-tracking system which has been used in films, virtual reality, and for research follows the objects. Theres also an RGB-D camera, which adds depth information to captured video.

The key was building modular objects. The uniform central pieces, made from aluminum, look like four-pointed stars and weigh about 100 grams. Each central piece contains markers on its center and points, so the Vicon system can detect its pose within a millimeter.

Smaller pieces in four shapes concave, triangular, rectangular, and circular can be magnetically attached to any side of the central piece. Each piece weighs between 31 to 94 grams, but extra weights, ranging from 60 to 150 grams, can be dropped into little holes in the pieces. All pieces of the puzzle-like objects align both horizontally and vertically, which helps emulate the friction a single object with the same shape and mass distribution would have. All combinations of different sides, weights, and mass distributions added up to 250 unique objects.

For each push, the arm automatically moves to a random position several centimeters from the object. Then, it selects a random direction and pushes the object for one second. Starting from where it stopped, it then chooses another random direction and repeats the process 250 times. Each push records the pose of the object and RGB-D video, which can be used for various video-prediction purposes. Collecting the data took 12 hours a day, for two weeks, totaling more than 150 hours. Humans intervention was only needed when manually reconfiguring the objects.

The objects dont specifically mimic any real-life items. Instead, theyre designed to capture the diversity of kinematics and mass asymetries expected of real-world objects, which model the physics of the motion of real-world objects. Robots can then extrapolate, say, the physics model of an Omnipush object with uneven mass distribution to any real-world object with similar uneven weight distributions.

Imagine pushing a table with four legs, where most weight is over one of the legs. When you push the table, you see that it rotates on the heavy leg and have to readjust. Understanding that mass distribution, and its effect on the outcome of a push, is something robots can learn with this set of objects, Rodriguez says.

Powering new research

In one experiment, the researchers used Omnipush to train a model to predict the final pose of pushed objects, given only the initial pose and description of the push. They trained the model on 150 Omnipush objects, and tested it on a held-out portion of objects. Results showed that the Omnipush-trained model was twice as accurate as models trained on a few similar datasets. In their paper, the researchers also recorded benchmarks in accuracy that other researchers can use for comparison.

Because Omnipush captures video of the pushes, one potential application is video prediction. A collaborator, for instance, is now using the dataset to train a robot to essentially imagine pushing objects between two points. After training on Omnipush, the robot is given as input two video frames, showing an object in its starting position and ending position. Using the starting position, the robot predicts all future video frames that ensure the object reaches its ending position. Then, it pushes the object in a way that matches each predicted video frame, until it gets to the frame with the ending position.

The robot is asking, If I do this action, where will the object be in this frame? Then, it selects the action that maximizes the likelihood of getting the object in the position it wants, Bauza says. It decides how to move objects by first imagining how the pixels in the image will change after a push.

Omnipush includes precise measurements of object motion, as well as visual data, for an important class of interactions between robot and objects in the world, says Matthew T. Mason, a professor of computer science and robotics at Carnegie Melon University. Robotics researchers can use this data to develop and test new robot learning approaches that will fuel continuing advances in robotic manipulation.

More:

Pushy robots learn the fundamentals of object manipulation - MIT News

Will you be willing to give your face to be put on thousands of robots worldwide for Rs 91,02,179 (100,000 pounds)? No, we are not just making this up. A start-up tech company is actually looking for a 'kind and friendly' face to put on their robots.

The unnamed firm is looking for human faces for the robots, which will be used as virtual friends for elderly people, and for the same, they have lodged a request with the company Geomiq.com.

Geomiq wrote that they were approached by a robotics company to help with the finishing touches to a 'state-of-art humanoid robot.' They did not give out a lot of details about the company due to a non-disclosure agreement that they have signed.

The company knows that it is an unusual request and that it is a big deal, which is why they are ready to pay so much money. The robots are set to go into production next year.

People are a little concerned about this demand and are asking why can't the company just make a new face exactly like Sophia, the social humanoid robot developed by Hong Kong based company Hanson Robotics.

Technology reporter Rowland Manthorpe took to his Twitter account to call it a bad sci-fi plot. He wrote, "Bad sci-fi plot alert. This unnamed company is looking for someone to be the face of its robot, which will be used as a 'virtual friend' for elderly people. It's offering 100,000 for the rights to your face in perpetuity."

Another Twitter user wrote, "Even I wouldn't have my face "in perpetuity" to which a person responded saying, "Who would?! I grew a beard to hide mine."

Would you be willing to give your face?

ALSO READ | World's first robot citizen gives wake-up call to the world on climate change

ALSO READ | The Girl From the Future

ALSO SEE | Meet Cosmopolitan India's newest cover girl, Sophia the robot

ALSO WATCH | Hanson Robotics Close to Making 'Truly Alive' Artificial Intelligence Robot

View post:

A company will pay you Rs 91 lakh to put your face on its robots. Internet is aghast - India Today

In a "Bad-sci-fi plot", a tech company is looking for a 'kind and friendly' face to be put on thousands of robots worldwide.

The unnamed tech company has even offered to pay around rupees 91 lakhs (100,000 pounds) to anyone, who agrees to have his face masked on these robots, which are being designed to act as "virtual friends" to elderly people.

The "unusual request" was forwarded to Genuique.com, a mechanical/industrial manufacturing firm, by the privately-funded robotics company as they are looking for the 'finishing touch' to the "state-of-art humanoid robot".

Genuique.com stated, "This is not our usual remit of request, which is why were making this public appeal to try and find the right person. The designer knows that this is a big deal, and has agreed a fee of 100,000 to license the rights to the right face."

The details of the company were not put out due to a non-disclosure agreement. However, the anonymity is also preserved due to the secretive nature of the project.

In a post shared on Twitter, Rowland Manthrope, a technology correspondent said that it's a bad-sci-fi plot".

Signing up to have your face put on robots around households is indeed an "extremely big decision" and this didn't go down well with Twitterati.

Get the best of News18 delivered to your inbox - subscribe to News18 Daybreak. Follow News18.com on Twitter, Instagram, Facebook, Telegram, TikTok and on YouTube, and stay in the know with what's happening in the world around you in real time.

';$('#first-watch-box').html(response+titleHeading);}},error: function(xhr, ajaxOptions, thrownError) {console.log('Something went wrong..');}});} var playWatchVideos = false;$(window).scroll(function() {var ividFirstScroll = $('.alsowatch').offset().top - 100;var tagsScroll = $('.tag').offset().top;var topOffsetIvid = $(window).scrollTop();var topIviddistance = topOffsetIvid - ividFirstScroll;var finalScrollEnd = ividFirstScroll + 650;var holaPlayDivId = $('div.video-js').attr('id'); var holaPlayerObj = videojs(holaPlayDivId);if(topOffsetIvid>ividFirstScroll && topOffsetIvid

See the article here:

Company Offers Nearly 90 Lakh Rupees If You Sign Up to Have Your Face on Robots - News18

PEKAN: IN a move to spur creative and innovative thinking among their pupils, teachers at SK Seri Terentang here decided to spend their own money to buy robotics kit sets early last year.

The school then made their debut at the International Robotics Competition in Putrajaya in February last year and unexpectedly bagged the runners-up prize.

On Oct 20, it made the country proud after winning a gold medal at the The Maker Robot Festival, a robotics competition which attracted participants from around the world in Ulsan, South Korea.

SK Seri Terentang Robotics Club adviser Mazlan Muhamad Yusof said the experience was an amazing journey, especially as the school exposed the pupils to robotics and programming only last year.

When we first started, we collected money from the teachers to buy robotics kit sets. We never expected the school team to make it this far.

Due to this, he said, the school now had its own Robotics Club where its members assembled up to six sets of robots and conducted programming activities every week.

We hope the achievements will encourage more Year One and Year Two pupils to join the club.

We have a limited number of robotics kit sets as they are costly. This is compounded by the fact that there is no special fund for the club.

However, club members are eager to learn and they spend long hours working on the robots, he said, adding that the pupils parents had financially contributed to their childrens South Korea trip.

Mazlan, who led the team of five pupils, along with two other teachers Roslaini Abdullah and Roszita Ajmain, said the decision to travel to South Korea was to give the participants exposure on robotics and Korean culture.

A Malaysian teacher who is in South Korea for a teacher exchange programme had told us about the festival. We decided to travel there hoping that the pupils could gain knowledge on robotics and learn a new culture.

Surprisingly, we emerged tops and won gold in the sumo-bot category, which is where two robots attempt to push each other out of a circle, he said.

The five pupils who travelled to the festival were Muhammad Imran Mazlan, 10, Muhammad Baqir Borhan, 12, Muhammad Danish Fawwaz Mohamad, 11, Nurul Iman Mazlan, 11, and Nur Ishamina Amni Mohd Imeran, 10.

On their next competition, Mazlan said the team would travel to Guangzhou, China, next month to compete in the 2019 MakeX Starter City Guardian competition. He said the schools team was picked to represent Malaysia after winning gold at the MakeX Malaysia Robotics Competition, which was held in Universiti Kebangsaan Malaysia in Bangi on Oct 12 and 13.

The participants have started their training and are making preparations for the competition. Training sessions include modifying robots and programming them to carry out instructions according to the requirements of the competition, he said, adding that some 60 countries would compete in the competition from Nov 28 to Dec 1.

SK Seri Terentang along with SK Putrajaya Presint 11(3) in Putrajaya recently won five medals including a gold medal in the South Korean robot festival.

Excerpt from:

Pekan school robotics team beats odds to win gold - New Straits Times

The SAM100 bricklaying robot at the Brighton Health Center South site of the University of Michigan Hospitals and Health Centers. Source: Construction Robotics

Until recently, construction was one of the least digitized and automated industries in the world. Many projects could be completed more efficiently with the help of the right construction robotics, mainly because the related tasks are incredibly repetitive.

While manual labor will likely always be a huge component of modern construction, technology has been steadily improving since the first pulleys and power tools. Robots, drones, autonomous vehicles, 3D printing, and exoskeletons are beginning to help get the work done. With low U.S. unemployment and shortages of skilled labor, automation is key to meeting demand and continued economic growth.

Construction robots may be involved in specific tasks, such as bricklaying, painting, loading, and bulldozing. We expect hundreds of AMRs in the next two years, mainly doing haulage, said Rian Whitton, an analyst at ABI Research. These robots help to protect workers from a hazardous working environment, reduce workplace injuries, and address labor shortages.

Many potential solutions rely on artificial intelligence and machine learning to deliver unprecedented levels of data-driven support. For instance, a driverless crane could transport materials around a worksite, or an aerial drone could gather information on a worksite to be compared against the plan.

Here are just a few examples of how robotics is transforming construction.

An example of how construction robotics are revolutionizing the industry can be seen in the HadrianX bricklaying machine from Australia-based FBR Ltd. (also known as Fastbrick Robotics). It employs an intelligent control system aided by CAD to calculate the necessary materials and movements for bricklaying.

Hadrian also measures environmental changes, such as movement caused by wind or vibrations, in real time. This data is then used to improve precision during the building process.

While Hadrian does require the use of proprietary blocks and adhesive, FBR noted that the related materials are 12 times bigger than standard house bricks and are lighter, stronger, and more environmentally sustainable.

Robots like Hadrian and SAM100 from Victor, N.Y.-based Construction Robotics promise to reduce operating costs and waste, as well as provide safer work environments and improve productivity. Hadrian can build the walls of a house in a single day, which is much faster than conventional methods.

While the major automakers and technology companies are working on self-driving cars, autonomous vehicles are already part of construction robotics.

Such equipment can transport supplies and materials. For instance, Volvo has been working on its HX2, an autonomous and electric load carrier that can move heavy loads without additional input. It has no driver cab and instead uses a digital logistics-driven control technology backed by what Volvo calls a vision system to detect humans and obstacles while on the move.

Another company, Built Robotics, which last month raised $33 million, offers autonomous bulldozers and excavators. AI guidance systems direct the equipment to their destinations and ensure that the necessary work is completed safely and accurately.

Autonomous vehicles and construction robotics is not intended to replace manual labor entirely, but to augment and enhance efficiency. Safety is vastly improved as well, as we eliminate the potential for human error.

Construction robotics and drones using sensors such as lidar with Global Positioning System technologies can provide vital information about a worksite. Along with AI, it can help predict what tasks are required.

Doxel Inc. makes a small tread-based robot that does exactly that. It scans and assesses the progress of a construction project by traversing the site. The information it collects is used to detect potential errors and problems early.

Doxels data is stored in the cloud, where its filtered through a deep-learning algorithm to recognize and assess more minute details. For example, the system might point out that a ventilation duct is installed incorrectly, and the early detection can allow for the proper correction well before costly revisions are needed.

Humans are still in the loop for much of construction robotics, combining the strengths of human supervision with multiple technologies. The Internet of Things, additive manufacturing, and digitization are contributing to the industrys growth, noted Caterpillar.

Painting drones are an excellent example, since they can be controlled via tablet or smartphone via an app, and they can report on the data they gather thats analyzed in the cloud.

Remote-control technology can also be applied to semi-autonomous vehicles. Project managers can use it to deliver instructions and orders to their workforce instantly.

Barcelona-based Scaled Robotics offers construction robotics that can be remotely controlled by mobile devices. The companys Husky unmanned ground vehicle can roam a construction site and capture critical information via multiple sensors. The data is transferred to the cloud, where its used for building information modeling (BIM) of the project.

Before, during, and after a construction project, many assessments require the review of a worksite and surrounding area. Limited surveillance is also necessary for supervising workers and securing the site. In addition, project managers and supervisors must walk the site to conduct final inspections. Construction robotics and drones can help all of these processes.

Aerial drones and ground-based robots can survey a worksite and gather multiple types of data, depending on the sensors used. Augmented reality and virtual reality can enable operators to get a realistic and real-time feel for what the drones are seeing.

While donning a VR headset, for instance, viewers can see a live feed of captured video from the drone. More importantly, that immersive experience is provided remotely, so project managers dont even have to be on the job site to get an accurate assessment. The video feed is also recorded for playback at a later time, providing yet another resource.

Companies are already using drone technology to this end. In 2018, Chinese drone maker DJI announced a global partnership with Skycatch for a fleet of 1,000 high-precision custom drones to create 3D site maps and models of project sites.

The global market for construction robotics also represents a huge opportunity for developers and suppliers. It could grow from $22.7 million in 2018 to $226 million by 2025, predicts Tractica. Research and Markets estimates that the market will grow to $126.4 million by 2025.

According to the International Federation of Robotics and the Robotic Industries Association, the construction robotics market will experience a compound annual growth rate (CAGR) of 8.7% between 2018 and 2022. Research firm IDC is more bullish, predicting a CAGR of 20.2%.

Automation and digitization are driving a revolution in the construction industry, which has historically been slow to adopt new technologies. From design through final inspection and maintenance, the full benefits of construction robotics have yet to be realized.

Originally posted here:

Construction robotics is changing the industry in these 5 ways - Robot Report

Find all our Student Opinion questions here.

Last week, a robotic hand successfully solved a Rubiks Cube. While that feat might seem like a fun parlor trick, its a sign that robots are being programmed to learn and not just memorize.

Robots are already playing important roles inside retail giants like Amazon and manufacturing companies like Foxconn by completing very specific, repetitive tasks. But many believe that machine learning will ultimately allow robots to master a much wider array of more complex functions.

When you imagine the future of robots and artificial intelligence, do you get excited? Do you envision a world of benefits for humankind? Or does an automated future fill you with concern and fear?

In If a Robotic Hand Solves a Rubiks Cube, Does It Prove Something? Cade Metz writes about how this five-fingered feat could show important progress in A.I. research:

Last week, on the third floor of a small building in San Franciscos Mission District, a woman scrambled the tiles of a Rubiks Cube and placed it in the palm of a robotic hand.

The hand began to move, gingerly spinning the tiles with its thumb and four long fingers. Each movement was small, slow and unsteady. But soon, the colors started to align. Four minutes later, with one more twist, it unscrambled the last few tiles, and a cheer went up from a long line of researchers watching nearby.

The researchers worked for a prominent artificial intelligence lab, OpenAI, and they had spent several months training their robotic hand for this task.

Though it could be dismissed as an attention-grabbing stunt, the feat was another step forward for robotics research. Many researchers believe it was an indication that they could train machines to perform far more complex tasks. That could lead to robots that can reliably sort through packages in a warehouse or to cars that can make decisions on their own.

The article continues:

A robot that can solve a Rubiks Cube is not new. Researchers previously designed machines specifically for the task devices that look nothing like a hand and they can solve the puzzle in less than a second. But building devices that work like a human hand is a painstaking process in which engineers spend months laying down rules that define each tiny movement.

The OpenAI project was an achievement of sorts because its researchers did not program each movement into their robotic hand. That might take decades, if not centuries, considering the complexity of a mechanical device with a thumb and four fingers. The labs researchers built a computer system that learned to solve the Rubiks Cube largely on its own.

What is exciting about this work is that the system learns, said Jeff Clune, a robotics professor at the University of Wyoming. It doesnt memorize one way to solve the problem. It learns.

Students, read the entire article, then tell us:

How do you feel about the future of robots, artificial intelligence and automation? Do you envision a world of progress? Or dystopian peril?

Do you fear countless professions becoming obsolete, the merging of humans and machines, an age of artificial intimacy or a potential robot uprising? Or are you excited about all the possibilities such a future could hold? Explain why you feel the way you do.

Researchers have already built machines to solve a Rubiks Cube in under a second. And even though OpenAIs hand can solve the puzzle in as little as four minutes, it drops the cube eight times out of 10. In your opinion, how big a deal is this robotic hand demonstration? Why do you think it made the news?

Students 13 and older are invited to comment. All comments are moderated by the Learning Network staff, but please keep in mind that once your comment is accepted, it will be made public.

Read the original:

Does the Future of Robots Get You Excited, or Fill You With Dread? - The New York Times

If youre at a desk with a pen or pencil handy, try this move: Grab the pen by one end with your thumb and index finger, and push the other end against the desk. Slide your fingers down the pen, then flip it upside down, without letting it drop. Not too hard, right?

But for a robot say, one thats sorting through a bin of objects and attempting to get a good grasp on one of them this is a computationally taxing maneuver. Before even attempting the move it must calculate a litany of properties and probabilities, such as the friction and geometry of the table, the pen, and its two fingers, and how various combinations of these properties interact mechanically, based on fundamental laws of physics.

Now MIT engineers have found a way to significantly speed up the planning process required for a robot to adjust its grasp on an object by pushing that object against a stationary surface. Whereas traditional algorithms would require tens of minutes for planning out a sequence of motions, the new teams approach shaves this preplanning process down to less than a second.

Alberto Rodriguez, associate professor of mechanical engineering at MIT, says the speedier planning process will enable robots, particularly in industrial settings, to quickly figure out how to push against, slide along, or otherwise use features in their environments to reposition objects in their grasp. Such nimble manipulation is useful for any tasks that involve picking and sorting, and even intricate tool use.

This is a way to extend the dexterity of even simple robotic grippers, because at the end of the day, the environment is something every robot has around it, Rodriguez says.

The teams results are published today in The International Journal of Robotics Research. Rodriguez co-authors are lead author Nikhil Chavan-Dafle, a graduate student in mechanical engineering, and Rachel Holladay, a graduate student in electrical engineering and computer science.

Physics in a cone

Rodriguez group works on enabling robots to leverage their environment to help them accomplish physical tasks, such as picking and sorting objects in a bin.

Existing algorithms typically take hours to preplan a sequence of motions for a robotic gripper, mainly because, for every motion that it considers, the algorithm must first calculate whether that motion would satisfy a number of physical laws, such as Newtons laws of motion and Coulombs law describing frictional forces between objects.

Its a tedious computational process to integrate all those laws, to consider all possible motions the robot can do, and to choose a useful one among those, Rodriguez says.

He and his colleagues found a compact way to solve the physics of these manipulations, in advance of deciding how the robots hand should move. They did so by using motion cones, which are essentially visual, cone-shaped maps of friction.

The inside of the cone depicts all the pushing motions that could be applied to an object in a specific location, while satisfying the fundamental laws of physics and enabling the robot to keep hold of the object. The space outside of the cone represents all the pushes that would in some way cause an object to slip out of the robots grasp.

Seemingly simple variations, such as how hard robot grasps the object, can significantly change how the object moves in the grasp when pushed, Holladay explains. Based on how hard youre grasping, there will be a different motion. And thats part of the physical reasoning that the algorithm handles.

The teams algorithm calculates a motion cone for different possible configurations between a robotic gripper, an object that it is holding, and the environment against which it is pushing, in order to select and sequence different feasible pushes to reposition the object.

A new algorithm speeds up the planning process for robotic grippers.A robot in the lab is shown picking up a block letter, T, and pushing it against a nearby wall to re-angle it, before setting it back down in an upright position.

Its a complicated process but still much faster than the traditional method fast enough that planning an entire series of pushes takes half a second, Holladay says.

Big plans

The researchers tested the new algorithm on a physical setup with a three-way interaction, in which a simple robotic gripper was holding a T-shaped block and pushing against a vertical bar. They used multiple starting configurations, with the robot gripping the block at a particular position and pushing it against the bar from a certain angle. For each starting configuration, the algorithm instantly generated the map of all the possible forces that the robot could apply and the position of the block that would result.

We did several thousand pushes to verify our model correctly predicts what happens in the real world, Holladay says. If we apply a push thats inside the cone, the grasped object should remain under control. If its outside, the object should slip from the grasp.

The researchers found that the algorithms predictions reliably matched the physical outcome in the lab, planning out sequences of motions such as reorienting the block against the bar before setting it down on a table in an upright position in less than a second, compared with traditional algorithms that take over 500 seconds to plan out.

Because we have this compact representation of the mechanics of this three-way-interaction between robot, object, and their environment, we can now attack bigger planning problems, Rodriguez says.

The group is hoping to apply and extend its approach to enable a robotic gripper to handle different types of tools, for instance in a manufacturing setting.

Most factory robots that use tools have a specially designed hand, so instead of having the abiity to grasp a screwdriver and use it in a lot of different ways, they just make the hand a screwdriver, Holladay says. You can imagine that requires less dexterous planning, but its much more limiting. Wed like a robot to be able to use and pick lots of different things up.

This research was supported, in part, by Mathworks, the MIT-HKUST Alliance, and the National Science Foundation.

Read the original post:

Giving robots a faster grasp - The MIT Tech

Howie Choset speaks at RoboBusiness 2019.

October 17, 2019John K. Waters

As both a robotics professor and robotics entrepreneur, Howie Choset has seen the pros and cons of manufacturing research and manufacturing businesses. Somewhere between the two lies a Valley of Death that many companies fall into as they try to bring their research to a profitable business. Finding ways to bridge that gap was one theme during Chosets keynote session at this years RoboBusiness 2019 conference.

The differences between the two worlds are rather stark, Choset explained. In academia, its all about creativity and big breakthroughs; in industry, its about incremental advances in products that sell. In academia, theres little fear of risk, because the work is grant-based, and incremental is almost an insult; in industry, investors expect profits, and failure actually hurts.

Its also true that robotics innovations in general havent drifted over the valley from academia to industry as easily as, for example, the technologies behind self-driving cars, he said.

Yet the payoff for investors in this technology has actually been quite steady. The industrial robotics business, for example, has sustained a growth rate of around 20% over the past few years, Choset noted.

That might seem small to an investor looking for a big return, he said, but the truth is, the robotics community at large is selling more industrial robots each year than we did the year before. Its not the kind of growth that would excite a standard VC style investor, which is why I think strategic partners make a lot more sense for robotics companies. It did for the companies that I started. Our main investors were generally strategic partners, who saw additional value beyond the payoff they could get from buying and selling this company.

Choset has been straddling these two worlds for much of his career. In addition to serving as the co-director of the Biorobotics Lab and director of the Robotics Major at Carnegie Melon, he has collaborated with his students to form several companies, including Medrobotics, for surgical systems, Hebi Robotics, for modular robots, and Bito Robotics, for autonomous guided vehicles. Choset also co-led the formation of the Advanced Robotics for Manufacturing Institute.

Howie Choset discussed the differences between academia and the robotics industry.

Choset said he worries about the impact of the current lack of major industrial robot manufacturing in the U.S. And we should be worried, too, he said. He shared a list of the most common industrial robots in use today, none of which are made in the U.S.

The Europeans and the Asians are the ones making these robots right now, he said, and we need to create an ecosystem where we can be at least on par with them, if not ahead. And I do believe this is an issue of national security.

Among the challenges he foresees for robotics manufacturing is something Choset called mass production in quantities of one.

We are going to expect that each one of our cars, for example, is going to be different, he said. Boeing will be the first to tell you that no two airplanes are really ever going to be the same. And increasing demand for ever faster delivery is already ratcheting up in the logistics space, he said. Especially within supply logistics, were going to have to be more nimble, because this is what the market is demanding; this is what our competitors and our peers are going to be able to deliver.

Choset shared videos of Hebis modular robotic snake, the surgical version of which is FDA-approved and currently in use in the U.S. and Europe. He foresees this kind of modularity as a powerful enabler in the robotics industry. Our goal was to make building robots good robots, almost industrial strength as easy as if youre playing with Legos, he said. The modular robots attracted the attention of the oil and gas industry, he said, and Chevron is adapting them for pipeline inspection.

There is an opportunity for innovation in manufacturing, which hasnt evolved much since the 1980s in many facilities, when it comes to soft changes. The installed technology represents huge investments that you cant just rip and replace, but the software can be reprogrammed to provide new, more customized capabilities for the hardware.

Of course, there is this trend in manufacturing right now toward cooperatives, or collaborative robots, he added, which are doing quite well. Im optimistic about this trend, and I think its something we also should be embracing.

Another technology Choset threw a spotlight on during his talk was mobile manipulators. His company, Hebi, and Canadas Clearpath Robotics have been collaborating. With mobile manipulators, you now have the opportunity to bring parts to the work, as opposed to the work coming to the parts, he explained. In other words, once you lay down an assembly line, youre pretty much fixed to that architecture. But if you have more independent agents running around the factory floor, bringing parts to the work at the right time, you create the opportunity now to decrease the time that it takes to set up your manufacturing facility. And you also have the opportunity to customize that facility and change things on the fly.

Choset also talked about a push in the industry for universal interfaces. The robotics industry needs universal interfaces, he said, because its not possible right now to mix and match robots in manufacturing settings. Were now in a period thats similar to where we were with PCs a decade or two ago, he said. The advent of standards in that industry facilitated incredible product improvements.

Among the challenges Choset sees for the industry in the near future: growing demand for robots that can manipulate soft objects, robots to assist workers in confined spaces and ergonomically challenging environments, and automated food preparation with a focus on reducing waste.

I think were going to see robots interacting more with people than we have in the past, Choset said at the end of his talk. Right now, we define collaboration as simply making it possible for a human to work in close proximity to a robot without getting hurt. But that idea, in particular is evolving. Were increasingly see more of a spectrum of models, where the robot and the person will be playing to each others strengths.

To see Chosets presentation slides at RoboBusiness 2019, click here.

See the rest here:

How to Bridge the 'Valley of Death' in Manufacturing Automation - Robotics Business Review