From Ian Lenz, Honglak Lee, Ashutosh Saxena:
We consider the problem of detecting robotic grasps in an RGB-D view of a scene containing objects. In this work, we apply a deep learning approach to solve this problem, which avoids time-consuming hand-design of features. This presents two main challenges. First, we need to evaluate a huge number of candidate grasps. In order to make detection fast and robust, we present a two-step cascaded system with two deep networks, where the top detections from the first are re-evaluated by the second. The first network has fewer features, is faster to run, and can effectively prune out unlikely candidate grasps. The second, with more features, is slower but has to run only on the top few detections. Second, we need to handle multimodal inputs effectively, for which we present a method that applies structured regularization on the weights based on multimodal group regularization. We show that our method improves performance on an RGBD robotic grasping dataset, and can be used to successfully execute grasps on two different robotic platforms... (homepage) (full pdf paper)
From french quad racing association Airgonay:
From Harvard Biodesign Lab:
The Soft Robotics Toolkit is a collection of shared resources to support the design, fabrication, modeling, characterization, and control of soft robotic devices. The toolkit was developed as part of educational research being undertaken in the Harvard Biodesign Lab. The ultimate aim of the toolkit is to advance the field of soft robotics by allowing designers and researchers to build upon each other’s work. The toolkit includes an open source fluidic control board, detailed design documentation describing a wide range of soft robotic components (including actuators and sensors), and related files that can be downloaded and used in the design, manufacture, and operation of soft robots. In combination with low material costs and increasingly accessible rapid prototyping technologies such as 3D printers, laser cutters, and CNC mills, the toolkit enables soft robotic components to be produced easily and affordably... (project's homepage)
From Evan Ackerman at IEEE Spectrum:
The video below has four parts to it: the first shows the difference between the robotic octopus swimming with just flexible arms, and swimming with just flexible arms in addition to a web. The most obvious difference is the speed: just over 100 millimeters per second with arms only, and up to 180 mm/s (or 0.5 body lengths per second) with the web. This is a significant increase, obviously, but what's more important is the overall cost of transport (CoT), which is a measure of the efficiency of the robot (specifically, the ratio of the energy put in over the resulting speed). The CoT for the arms-only version is 0.85, whereas the web drops that down to 0.62. So yeah, having that web in there is better in almost every way... (cont'd)
From Nixie's homepage:
Nixie is a tiny wearable camera on a wrist band. The wrist straps unfold to create a quadcopter that flies, takes photos or video, then comes back to you... (cont'd)
From Clive Thompson:
A few weeks ago I got duped by a robot. In the mail.
I was sifting through my dead-tree postal mail and tossing junk in the recycling bin. Nearly everything that arrives in my mailbox is junk, so I was tossing, tossing, tossing … until suddenly, whoops: A hand-addressed letter. This looked legit, so I ripped it open — only to find it was an oily invitation to take out a second mortgage on my home. I’d been fooled... (cont'd)
From IEEE Spectrum:
The airplane is initially parked on a runway of an airport. The robot prepares the flight by 1) pulling throttle to zero-point, 2) turning on the battery, 3) the altimeter, 4) the avionics, 5) the fuel pump, and 6) start the engine while pressing the switches on the panel. Then, PIBOT grabs the two control sticks for flight control and brakes are released. When the heading of the airplane aligns with the runway within an error less than 5 degree and its speed exceeds the taxiing speed, the second sequence begins and PIBOT increase the power... (cont'd)
Indigogo campaign for Pawly:
Take your playtime to the next level with Pawly's accessory. Pawly can be equipped to play and reward your pets in real time, mimicking the way pet-owners would play with their pets.
Reward your pet when they do back flips when you're away. Toss them a treat with Pawly's Treat Blaster. This safe but exciting accessory will shoot out a treat at the press of a button. The LEDs found on the dome light up, followed by a sound before shooting out their favourite treats.
To use the Treat Blaster, mount it on top of Pawly by lining up the teeth of the accessory to the three holes on top of Pawly. Turn on Pawly's app and start blasting away.. (cont'd)
Kickstarter for version 4 of ArduIMU:
Initially, the ArduIMU project was started as an open source project by 3DRobotics in 2007 to create an inertial measurement unit based on the Arduino™. We contributed to the software development of that project, but, the initial ArduIMU was meant to be used solely as an inertial measurement unit. We want to do better than that; so we developed a brand new platform with other sensors such as a barometer, relative humidity sensor, and light sensor. We also included wireless communication capabilities as well an SD card for data logging and storage. Since then, we have gone through many revisions, adding even more sensors and functions which are present in the latest ArduIMU V4. With this new augmented and improved sensor board we are redefining the term IMU. We proudly present our Arduino™ based Integrated Measurement Unit: the ArduIMU V4... (cont'd)
Further info from Evan Ackerman at IEEE Spectrum:
We were wrong: it's not running untethered, it's bounding untethered. And unconstrained. And outdoors!
Two things strike us as particularly amazing about this: the first thing is that it's quiet, powered by electric motors and batteries. We've come to expect that compact systems capable of delivering high amounts of power rely on liquid fuels and hydraulics, because that's how you get the most power density: it's why Boston Dynamics uses gasoline engines to power hydraulic pumps on all of its dynamic robots. Also, high torque electric motors (like you'd need to get a robot to jump) have a tendency to overheat and destroy themselves, but MIT seems to have solved all of these issues, since they have a bounding, battery-powered robot that works. We're not sure yet how long it works for, but it works... (cont'd)
YuMi is a human-friendly dual arm robot designed for a new era of automation, for example in small parts assembly, where people and robots work hand-in-hand on the same tasks. YuMi is short for ‘you and me,' working together.
YuMi has been developed to meet the flexible and agile production needs of the consumer electronics industry in the first instance. It will increasingly be rolled out to cover other market sectors. YuMi is a collaborative, dual arm assembly solution with the ability to feel and see. The robot's soft, padded dual arms, combined with innovative force-sensing technology ensure the safety of YuMi's human co-workers. Safety is built into the functionality of the robot itself so that it can work cage-free... (cont'd)
From Boston Herald:
A company with U.S. headquarters in Marlborough that was recently awarded FDA approval to sell its robotic exoskeletons for paraplegics plans to raise $50 million in an IPO this week, possibly on Friday.
Israeli-based ReWalk Robotics is planning to sell 3.5 million shares for between $14 and $16 each, which puts it at the low end of the 13 local health care companies which have gone public since the beginning of the year, more than any other year in history. Most of those have been biotech companies, however, making ReWalk the first robotics-focused company to do so in at least a couple of years... (cont'd)
James Dyson explaining Dyson's new 360 Eye robotic vacuum:
From University of Tokyo:
ACHIRES is composed of high-speed vision and high-speed actuators to achieve instantaneous recognition and behavior. The similar technologies are used in our Janken (Rock Paper Scissors) Robot. High-speed vision detects the state of the biped robot including the timing of landing at 600 fps. The biped mechanism with the leg length of 14 cm is set to run in the sagittal plane. At present, the running velocity reaches 4.2 km/h. Simple control based on high-speed performance of sensory-motor system enables the biped robot to stably run without falling, unlike computationally expensive ZMP-based control which is commonly used for balance. The aerial posture is recovered to compensate for the deviation from the stable trajectory using high-speed visual feedback.
We also address a task of somersaulting. While running, the robot takes a big swing with one foot and jumps. After takeoff, both legs are controlled to curl up for high-speed rotation in the air. ACHIRES is going to be improved to push the envelope while demonstrating various biped locomotion tasks... (cont'd)
From Nick McCrea at Toptal:
In this article, I’m going to describe the control scheme of my simulated robot, illustrate how it interacts with its environment and achieves its goals, and discuss some of the fundamental challenges of robotics that I encountered along the way... (full article)
Records 361 to 375 of 675