Christian Smith shows a demo of ball-catching using a custom robot arm at CAS.
The idea is to cover a large enough area in short enough time to be able to do this.
A 'large enough area' has been defined as a 0.6 m by 0.6 m window, based on the precision of a normal underhand ball toss. A 'short enough time' has been defined as 0.5 s, based on the flight time of ballistic objects and the reaction times of the teleoperated system.
In simulations the robot arm can, if started in the middle of the operating window, reach any point therein within 0.3 s, which should be fast enough and also impressive.
Seeing the video with a bad connection sometimes you wont even see the arm movement to catch the ball, that fast!
Complete documentation can be found here.
Lorenzo Jamone, Matteo Fumagalli and Giorgio Metta from the University of Genoa, work on a project called "Machine-Learning Based Control of a Human-like Tendon-driven Neck".
For this purpose they builded James, a robot able to orientate its head as needed by actuating its tendon-driven parallel redundant robotic neck.
In Lorenzo's YouTube channel you can watch other videos of James learning to reach and grab objects but for the whole paper you must look here.
marcimatz shows us the Virtual Robot Experimentation Platform simulating the Expliner robot. Expliner (which is commercially available from Hibot Corporation is able to inspect very high power lines while power is on, and also manages to overcome spacers and other obstacles on its way. The Expliner CAD model is courtesy of Hibot Corporation. You can watch the robot in action here.
Ph. D. Student Orges Furxhi from the University of Memphis is working to use radiowaves to create real-time imaging.
While working on previous methods he found a much easier way to do it. He is using radiowaves in terahertz frequencies and a device he constructed, a spining disc with holes driled on it, to measure the amount of energy passing through.
An idea that can easily become essential to security one day.
Victor Adrian Prisacariu and Ian Reid from Oxford University propose a system for human computer interaction via 3D hand movements, based on a combination of visual tracking and a cheap, off-the-shelf, accelerometer. They use a 3D model and region based tracker, resulting in robustness to variations in illumination, motion blur and occlusions. At the same time the accelerometer allows them to deal with the multimodality in the silhouette to pose function. They show the effectiveness of their solution with multiple real-world tests and demonstration scenarios.
Daniel Mellinger from the University of Pennsylvania describes one of his works as "developing controllers for aggressive maneuvers with quadrotor helicopters". In action this means tremendous maneuverability and insane control in turning.
Perching seems simple compared to the double flips and the amazingly precise flight through windows. A UAV showing things to come, enjoy.
David Resnick choose a rather complex project for his first-year project in the Arts Computation Engineering program at UC Irvine. The overall goal of Undulating Flux is to bring the participant into what psychologist Mihalyi Csikszentmihalyi termed a flow state. A flow state is described as a state of being wherein action follows upon action according to an internal logic that seems to need no consc...
Oluf Skov Nielsen and Emil Pedersen present their master's thesis project Brick laying mobile manipulator at Maersk McKinney Moller Institite of Syddansk Unviversity.
The objective of the project is to use a mobile manipulator to build a simple wall by enabling it to autonomously lay bricks. The project is limited to not include the use of mortar or other gluing material, thereby simpli...
Mark Czubin and Domien Nowicki present their project for Mobile & Pervasive Computing at Hasselt University.
Mounted on the glove is an Arduino board with an accelerometer, flex sensor and infrared led, the infrared led is seen by the wiimote, and is used for tracking finger motion. The flex sensor is used for shooting and the accelerometer for reloading, last the communication between glove and laptop is done through an XBee module. I can't wait to see this with a first person shooter game.