Preloader image
   
44
blog,paged,paged-150,satellite-core-1.0.6,unselectable,satellite-theme-ver-3.1,,smooth_scroll

Continuous 3D Sweep Matching with a Spinning 2D Laser

The vehicle trajectory is estimated using laser range scans generated by a SICK LMS291 rotating at 0.5Hz mounted on a skid-steer loader. Because the spinning laser requires one second to sweep a full 3D scan, the raw point clouds are highly distorted by the vehicle motion. The sweep-matching algorithm is able to continuously correct the motion-distorted point cloud to produce a 6DoF trajectory estimate without the use of any other sensors. Details of the algorithm are published by Michael Bosse and Robert Zlot in the IEEE International Conference on Robotics and Automation 2009 under the title “Continuous 3D Scan-Matching with…

Grasping manipulation and exploration tasks with the OctArm continuum manipulator

In the biological world, grasping, manipulation, and exploration tasks are frequently performed by tongues, trunks, and tentacles. Continuum robots such as the OctArm seek to mimic this wide range of abilities, allowing use in unstructured environments. Although a large variety of continuum robots have been developed as both commercial products and as research endeavors, video studies which concentrate on grasping, manipulation, and exploration tasks are rare. This video addresses this lack by illustrating the types of tasks for which continuum manipulators are uniquely suited. A series of outdoor and laboratory tests reveal the wide variety of novel grasping, manipulation, and…

Automatic Pick and Place of 40 Microns Silicon Parts

Automation of the robotic microassembly is currently (2008) a high stake to be able to produce assembled microsystems (MEMS). The video shows an automatic pick-and-place of 40×40 microns silicon parts in real time. The time cycle is about 1.8 seconds which is comparable with industrial assembly process at the scale of millimeter.  

Hand bot retrieve a book

The prototype of the hand-bot managed to successfully climb a shelf and retrieve a book autonomously. It achieved this thanks to its innovate synergy between its climbing and manipulation subsystems.  

IEEE R-AM 2011 – Hand bot climbing a shelf

See the first climb of a shelf performed by the hand-bot. The robot launches a rope and then uses it as the main lift force. The handbot keeps its balance using its two arms which have a total of seven degrees of freedom. The Handbot detects the shelf side using infrared sensors located in its hands. The movements of the robot are fully autonomous with no remote control behind.  

Image guided robot for skull base surgery

We developed a neurosurgical robotic system to address the challenges of neurosurgery. It uses a NeuroMate robot as a new and effective surgical tool by adding a force sensor to the tip of the robot. The basic motivation behind our project is to significantly reduce the operating time of a complex procedure—bone milling at the skull base—by providing a safer and more reliable surgical tool. The robotized solution would only be used for the removal of the bone tissue, to gain access to the soft tissue region affected by the disease. We use a preoperative image, such as CT, to…

Jaemi HUBO Humanoid robot

The Drexel University-led research team, DASL, unveiled the newest, most central member of its collaboration with a team of Korean researchers: Jaemi, a humanoid (HUBO). Jaemi HUBO embodies efforts to advance humanoid development and enhance the concept of human-robotic interaction. The project’s goal is to enable humanoids to interact with their environment, and enhancement plans include enabling the humanoid to move over rugged terrain, in unstructured environments and to interact socially with humans and handle objects.  

Large wire driven parallel rescue crane

The presented robot is a portable, fully autonomous wire-driven crane with 6 d.o.f. It can be deployed on site in about 10 min and has a lifting capacity of about 2 tons. This video shows also that this a communicating tool allowing doctors to get physiological information (pulse rate, type of injury) on the victims while the victim is being transferred. In the presented video the robot’s workspace is 20m X 20m x 12m  

Learning Motor Primitives for Robotics

This video exhibits current progress in motor skill learning. It shows two tasks: paddling a ball on a string and ball-in-a-cup, which is also known as Kendama. For paddling a ball on a string, a demonstration of a human teacher is shown and the system can directly acquire this behavior by imitation learning. For ball-in-a-cup, first an imitation is demonstrated by a teacher. Subsequently, the robot attempts to reproduce the behavior but fails. Hence, the robot explores new strategies getting better most of the time — selected trials are being shown in the video. After 90-100 trials, the robot finally…

OctArm a Soft Robotic Manipulator

Hitting and batting tasks, such as tennis forehands, ping-pong strokes, or baseball batting, can be learned using motor primitives Here, we first show how a robot is given several different forehands for imitation learning. After learning, the forehands are tried on first on a static target to show the natural smoothness and speed of the learned movement. Subsequently, they are shown to work even with balls shot by a ball gun.  

RoboCup @ Home 2009 Best of Innovation Award Winner NimbRo

The video shows highlights from the [email protected] competition, which took place in Graz, Austria, in July 2009. The [email protected] league aims to develop service and assistive robot technology with high relevance for future personal domestic applications. It is the largest international annual competition for autonomous service robots. A set of benchmark tests is used to evaluate the robots abilities and performance in a realistic non-standardized home environment setting.   In the video, the domestic service robots Robotinho and Dynamaid of Team NimbRo (University of Bonn, Germany) demonstrate their skills in robust indoor navigation, object manipulation, and intuitive multimodal human-robot interaction….

Robotic Microassembly of 40 Microns Silicon Parts

Robotic microassembly is a high stake to improve microsystems (MEMS) functionalities and modularity. The video shows a microassembly of 40×40 microns silicon parts. Piezoelectric microgripper has 4 Degree Of Freedom (DOF) – 2 for each fingers – and is placed on a 6 Degree Of Freedom (DOF) microrobot. Microgripper is used to handle the object, and to perform microassembly.  

error: Context Menu disabled!