Deformable object tracking is used in many roboticsapplications including biomanipulation, vision-based force sensing, and the control of deformable structures. A tracking algorithm that is robust to occlusions and t...
详细信息
Deformable object tracking is used in many roboticsapplications including biomanipulation, vision-based force sensing, and the control of deformable structures. A tracking algorithm that is robust to occlusions and to spurious edges is essential since these situations can arise unexpectedly in the unstructured environments in which robots must operate. This paper presents a deformable object tracking algorithm that is robust to occlusion and to spurious edges. Robust statistical methods are used to handle occlusion and a modification of the Canny edge detector is presented to handle spurious edges. The modification of the Canny edge operator makes use of information about the object being tracked in order to eliminate spurious edges. The deformable object tracking algorithm's performance is evaluated visually and quantitively by tracking a four degree-of-freedom compliant gripper.
Robot skills that are needed to operate completely autonomously in a real complex scenario populated by humans are still beyond the capabilities that the current robotic technology offers. In spite of that, in a varie...
详细信息
Robot skills that are needed to operate completely autonomously in a real complex scenario populated by humans are still beyond the capabilities that the current robotic technology offers. In spite of that, in a variety of assistant applications where human and robot are closely tied, mobile robots can perform an effective and valuable work if humans in the surroundings (or the assisted person) would be enabled to extend the robot abilities through skills either not supported by the machine or supported in a different and (maybe) less dependable manner. To achieve that, robot and human must closely interact and collaborate at all levels of the robotic architecture, including deliberation, control and execution. This paper proposes a new robotic architecture, called ACHRIN, which supports a strong integration of the human into the robotic system in order to improve the overall performance of the robot as well as its dependability. Our human-robot integration relies to a great extend on sharing symbolic concepts of the world (cognitive integration). ACHRIN has been implemented and tested in a real rehabilitation robot: a robotic wheelchair that provides mobility to impaired or elderly people.
The following topics are dealt with: image processing techniques; knowledge discovery and decision making; Web technologies and Web services; intelligent agents and ontologies; encryption and security e-commerce appli...
The following topics are dealt with: image processing techniques; knowledge discovery and decision making; Web technologies and Web services; intelligent agents and ontologies; encryption and security e-commerce applications; knowledge management and information management; robotics control; hybrid systems; personal assistant agents; fuzzy systems; industrial automation; data mining; information retrieval; risk analysis; multiagents systems; intelligent control systems; collision avoidance; fault diagnosis; signal processing; ant colony computation; genetic computation; evolutionary computation
Climbing robots can perform many tasks inaccessible to other robots or humans such as inspection, repair, cleaning, surveillance, and exploration. This paper presents and discusses the design, fabrication, and evaluat...
详细信息
Climbing robots can perform many tasks inaccessible to other robots or humans such as inspection, repair, cleaning, surveillance, and exploration. This paper presents and discusses the design, fabrication, and evaluation of two novel bio-inspired climbing robots. Both are inspired by the locomotion of Geckos, a highly skilled natural climber. They are developed for terrestrial and extra-terrestrial environments, and their kinematics is inspired by the Geckos’ gait. The first relatively large robot actuated by conventional motors is designed to operate at both in Earth and space scenarios. The second robot, whose motion is controlled using shape memory alloy actuators and size can be miniaturized to few centimeters scale, is designed for terrestrial applications. Preliminary prototypes of these robots are developed, demonstrated, and evaluated by steep and flat acrylic surface climbing tests. Current robots can successfully climb up to 65 ° slopes at 2 cm/sec speeds.
Sampling-based planners have solved difficult problems in many applications of motion planning in recent years. In particular, techniques based on the Rapidly-exploring Random Trees (RRTs) have generated highly succes...
详细信息
Sampling-based planners have solved difficult problems in many applications of motion planning in recent years. In particular, techniques based on the Rapidly-exploring Random Trees (RRTs) have generated highly successful single-query planners. Even though RRTs work well on many problems, they have weaknesses which cause them to explore slowly when the sampling domain is not well adapted to the problem. In this paper we characterize these issues and propose a general framework for minimizing their effect. We develop and implement a simple new planner which shows significant improvement over existing RRT-based planners. In the worst cases, the performance appears to be only slightly worse in comparison to the original RRT, and for many problems it performs orders of magnitude better.
Recent developments in the field of robotics, smart materials, micro actuators and mechatronics have opened a new frontier for innovation and development in millimeter scale actuators for use in medical robotics. In t...
详细信息
Recent developments in the field of robotics, smart materials, micro actuators and mechatronics have opened a new frontier for innovation and development in millimeter scale actuators for use in medical robotics. In this paper a novel design idea for developing a millimeter scale actuator is presented for actuating the end effector of a robot performing minimally invasive surgery (MIS). This actuator is designed by combining DC motor and shape memory alloy (SMA) actuator in series. The designed actuator is 5 mm in diameter and 40 mm in length and is used to actuate 10 mm long needle driver jaws, while generating a force of 15 N and a gripping force of 5.5 N.
It is a great challenge for a robot In space to track and capture a free-flying object in the future space operations. In previous research, most of them employed model-based method which requires robot model in advan...
详细信息
This paper describes the dynamic modeling of linear object deformation based on differential geometry coordinates. Deformable linear objects such as cables and strings are widely used in our daily life, electric indus...
详细信息
This paper describes the dynamic modeling of linear object deformation based on differential geometry coordinates. Deformable linear objects such as cables and strings are widely used in our daily life, electric industries, medical operations. Modeling, control, and manipulation of deformable linear objects are keys to many applications. We have proposed the differential geometry coordinates to describe the 2D/3D deformation of a linear object with the minimum number of parameters. Based on this description, we have formulated the static deformation of a linear object using the differential geometry coordinates but the dynamic deformation has not been investigated yet. In this paper, we apply differential geometry coordinates to the dynamic modeling of linear objects. First, we formulate the dynamic 2D deformation of an inextensible linear object based on a differential geometry coordinate system. Second, we show simulation results using the proposed modeling technique. Next, we apply the proposed dynamic modeling to the control of a flexible link.
This paper presents a new reinforcement learning algorithm for accelerating acquisition of new skills by real mobile robots, without requiring simulation. It speeds up Q-learning by applying memory-based sweeping and ...
详细信息
This paper presents a new reinforcement learning algorithm for accelerating acquisition of new skills by real mobile robots, without requiring simulation. It speeds up Q-learning by applying memory-based sweeping and enforcing the “adjoining property”, a technique that exploits the natural ordering of sensory state spaces in many robotic applications by only allowing transitions between neighbouring states. The algorithm is tested within an image-based visual servoing framework on a docking task, in which the robot has to position its gripper at a desired configuration relative to an object on a table. In experiments, we compare the performance of the new algorithm with a hand-designed linear controller and a scheme using the linear controller as a bias to further accelerate the learning. By analysis of the controllability and docking time, we show that the biased learner could improve on the performance of the linear controller, while requiring substantially lower training time than unbiased learning (less than 1 hour on the real robot).
暂无评论