From New York Times: It seemed like the perfect night life accessory for the South Beach set — an automated robotic parking garage where trendy clubgoers could park their Porsches with a futuristic touch of a button. Forget hiding your GPS and favorite Fendi sunglasses from a valet who might ding your new alloy wheels; this garage would park cars itself. Instead, malfunctions lasted for hours. Cars were smashed, and faulty machinery fell several stories to the ground. Sometimes vehicles were stuck for so long that garage operators had to pay for customers’ taxis... ( full story )
From BBC: Billion-dollar drone company DJI is expanding from consumer and camera drones into the agriculture industry. The Chinese firm's latest model is a crop-spraying drone, which it claims is "40 times more efficient" than manual spraying, despite having just 12 minutes of flight time. It will be released in China and Korea where hand-spraying is more common. DJI made $500m (£332m) in drone sales in 2014 and some analysts predict the firm will hit $1bn in sales this year. The Agras MG-1 has eight rotors and can carry up to 10kg of crop-spraying fluids per flight. The foldable device is also dustproof, water-resistant and made of anti-corrosive materials, the firm says on its website (in Chinese).
From MIT: Drake ("dragon" in Middle English) is a toolbox maintained by the Robot Locomotion Group at the MIT Computer Science and Artificial Intelligence Lab (CSAIL). It is a collection of tools for analyzing the dynamics of our robots and building control systems for them in MATLAB and C++, with a heavy emphasis on optimization-based design/analysis. Here is a quick summary of capabilities: Simulation Rigid-body dynamics including contact/collisions (hybrid+LCP) and kinematic loops Basic aerodynamics/fluid dynamics Sensor models (lidar, depth camera, imu, contact force/torque; cameras coming soon) Hand-derived models for many canonical control dynamical systems Easily add your own models/components Some support for stochastic models For all of the above we aim to expose sparsity and provide analytical gradients / symbolic analysis Primary limitations: code is optimized for analysis / planning / control design (as opposed to speed, generality)... ... Most of these models/tools are described in the companion textbook from an MIT course/MOOC . We've also recently started populating the Drake Gallery (contributions welcome!)... ( git repo )
From Mashable: That's right; online used car retailer Carvana has officially unveiled its first vending machine completely stocked with cars. Standing at five stories tall, the coin-operated (yes, really) glass tower vending machine can house as many as 20 used cars at a time. The facility also includes a welcome center and three customer delivery bays to which the cars are dispensed... ( full story )
By Silicon Valley Robotics via Robohub: The area of service robotics is getting active, with a new retail robot startup launching today. Tally is one of several robotics startups launching today at Haxlr8r’s 7th Demo Day. Tally is an inventory tracking robot platform fromSimbe Robotics and the “world’s first robotic autonomous shelf auditing and analytics solution” according to the press release. Tally is in trials with several North American retailers and will traverse aisles scanning and auditing merchandise to help stores maintain ideal product placement, fill inventory gaps, and find misplaced or mispriced items. Tally is also capable of autonomously returning to base to charge. “When it comes to the retail industry, shopper experience is everything. If a product is unavailable at the time the shopper wants to buy it, the retailer has missed an opportunity and disappointed their customer,” according to Brad Bogolea, CEO and Co-founder of Simbe Robotics. “Tally helps retailers address these challenges by providing more precise and timely analysis of the state of in-store merchandise and freeing up staff to focus on customer service.” Cont'd...
From Institute of Science and Technology Austria: Robotic systems controlled by a neural network spontaneously develop self-organized behaviors. Researchers propose a novel learning rule in PNAS to explain the development of sensorimotor intelligence. It is fascinating to observe a robot exploring its physical possibilities and surroundings, and subsequently developing different self-taught behaviors without any instructions. In their paper (DOI: 10.1073/pnas.1508400112) published on October, 26, 2015 in PNAS (Proceedings of the National Academy of Sciences), Professor Ralf Der from the Max Planck Institute for Mathematics in the Sciences, und Georg Martius, Postdoc and Fellow at the Institute for Science and Technology (IST Austria), demonstrate the emergence of sensorimotor intelligence in robots based on their proposed learning rule... ... To test their hypothesis, the authors use bioinspired robots consisting of a humanoid and a hexapod robot in physically realistic computer simulations. The robots receive sensory input from their bodies but are not given any form of instruction or task. What can then be observed is a rich spectrum of rhythmic behaviors of the robots as they explore various movements. Solely because of the tight coupling of environment, body, and brain (in this case an artificial neural network), the robots can obtain feedback from their situation and adapt quickly. This, together with a simple, learned self-model, allows them to develop a form of sensorimotor intelligence... ( full article ) ( paper ) ( videos and other materials )
From University of California, Irvine: VATIC is a free, online, interactive video annotation tool for computer vision research that crowdsources work to Amazon's Mechanical Turk. Our tool makes it easy to build massive, affordable video data sets and can be deployed on a cloud. After three years of research, vatic is now used by labs around the world to annotate the next generation of data sets. Features Crowdsource video annotation to Mechanical Turk Automatic quality assurance for good annotations Complex payment schedules and bonuses One-click worker compensation Optimized interface for video annotation Offline mode for expert annotator use (no MTurk) ( site ) ( github )
By JUN HONGO for The Wall Street Journal: Japan’s cabinet office, Kanagawa prefecture and Robot Taxi Inc. on Thursday said they will start experimenting with unmanned taxi service beginning in 2016. The service will be offered for approximately 50 people in Kanagawa prefecture, just south of Tokyo, with the auto-driving car carrying them from their homes to local grocery stores. According to the project organizers, the cabs will drive a distance of about three kilometers (two miles), and part of the course will be on major avenues in the city. Crew members will be aboard the car during the experiment in case there is a need to avoid accidents. Robot Taxi Inc., a joint venture between mobile Internet company DeNA Co. and vehicle technology developer ZMP Inc., is aiming to commercialize its driverless transportation service by 2020. The company says it will seek to offer unmanned cabs to users including travelers from overseas and locals in areas where buses and trains are not available. Cont'd...
From Thanos 6DOF Motion Simulator Electronics blog : Lately we see more and better DIY 6DOF platforms being build. The community is growing strong and the interest in 6DOF systems is bigger than ever. Its the only way to provide full immersion for motion simulation at the moment, that combined with other new technologies like the oculus rift or fully enclosed detailed cockpits can really make the difference. The times that desktop racers or flight enthusiasts were enjoying their simulation games or training for the real thing is past without return. Even older motion systems that provide 2DOF (seat movers) or 3DOF platforms are now the past, offering too little to the immersion of gaming... ( more projects )
From Alex Churchill: In the discussion on this site I assemble a Universal Turing Machine from Magic: the Gathering cards. But doesn't Magic involve the players making lots of choices? Normally, yes, it does. But occasionally in normal gameplay you get a sequence of three or four events in a row that are forced to happen by the cards and the rules of the game. The machine below just extends this idea to millions of forced choices in a row. The idea of my Magic Turing machine is that the players do nothing at all, except when the game offers them a choice. Once the in-game "machine" has started, processing continues without requiring any choices from the players, with one category of exceptions: Some of the cards in the machine say "You may [do X]. If you do, [Y happens]." In these cases, the machine arranges that the players will be able to do X, in precisely one way. It just requires the players to always choose to take the game up on any options they're offered... ( cont'd )
From Department of Mechanical and Process Engineering at ETH Zurich: The Aerial Construction project is a collaboration between the Institute for Dynamic Systems and Control and the Chair of Architecture and Digital Fabrication. The objective is to investigate and develop methods and techniques for robotic aerial construction... ( project homepage )
From 7Bot's Kickstarter campaign: In 2014, two of us co-founded project uArm with other two makers. There we received a lot of feedback from our Angel backers: More axis for more powerful applications, and more controlling dimensions (force control, speed control and flexible-joint). More rugged material rather than Acrylic, muscular servos hardly to be burned out. More intelligent API. Better inverse kinematics and path planning algorithms to make the movement more precise and smooth. More accessories and various of end-effectors. Our custom servos with precise position feedback allow you to quickly set it up and operate in teaching mode without any codes. In this mode, you can simply drag each joint of the robot to a serious of desired way points. The movements will be recorded, and could be replayed in an optimized path. A multi-platform supported 3D visualization application will be provided for you to manipulate the 7Bot Arm intuitively. With our 3D visualization application, you can easily set and read the position of each joint separately with real-time graphic interface. If you have two 7Bot Arms, you can build this amazing Humanoid robot -7Bot Arm Dual: Estimated shipping date is January 2015... (7Bot's Kickstarter campaign)
From uBreakiFix's Youtube channel:
Animated retelling of a Buckminster Fuller story and Ephemeralization (doing more with less):
Bot-maker Savioke announces an open-source wrapper for Intel's RealSense Camera, adding another low-cost 3D sensing solution to the roboticist's toolkit. The wrapper will allow developers to make use of the RealSense Camera, which enables robots to sense rich three-dimensional environments. "Intel RealSense Cameras bring great low-cost depth sensing to robotics, in a platform that is widely available and easy to integrate using ROS," says Steve Cousins, CEO of Savioke. Until recently, bot makers looking to incorporate 3D sensing on the cheap have relied on a sensor made by Israeli company PrimeSense. But in late 2013 PrimeSense was acquired by Apple for $350M, an indication of just how much potential the Cupertino-based giant sees in 3D sensing technology. Since the acquisition, robot developers have been eager for a flexible and cheap depth sensor. Intel, meanwhile, is making an aggressive move into the world of robotics, and the company was thrilled to offer ROS support for RealSense.
Records 76 to 90 of 121
MICROMO launches the new MC3/MCS motion control family. The new high performance, intelligent controllers are optimized for use with FAULHABER motors, offer electronics for simple operation with state-of-the-art interfaces for multi-axis applications, and provide a motion control system solution with the most compact integration into industrial grade housing.