Abstract-Nowadays dexterous manipulation of rigid objects using a robot hand can be achieved fairly well. However, grasping and manipulating deformable objects is still challenging as the force and tactile sensors which are commonly used in such applications can only provide local information about the deformation at the contact points. In this paper, a vision framework is proposed for 3D visually guided grasping and manipulation of deformable objects. This visual monitoring framework, which uses state-of-the-art computer vision methods, provides a robotic hand system with comprehensive monitoring of the deformable object that it manipulates as it tracks its deformation. Stereoscopic vision is used to detect and track in real time the deformation of non-rigid objects in three dimensions and within a complex environment. The technique is tested successfully in real robotic operation conditions using the Barrett hand. The actual object shape is rendered in the 3D virtual environment of the GraspIt! robotic simulator which also displays the hand configuration.
The paper discusses a novel unsupervised learning approach for tracking deformable objects manipulated by a robotic hand in a series of images collected by a video camera.
-This paper overviews the modern concepts adopted by cope with the inaccuracy of vision systems alone due to the robotics community during the past decade related to 3D
Designing a dexterous robotic hand able to interact intelligently with deformable objects constitutes a challenging area of research where many issues are yet to be solved. The complexity of such interactions requires the assistance of intelligent multisensory robotic systems that combine measurements collected from different sensors in order to accurately plan for the forces to be applied on the deformable object. This paper presents the development of a real-time multisensory robotic hand platform that incorporates live measurements of its internal position, velocity and force parameters along with data from external tactile sensors and a stereoscopic vision device. The resulting prototype of the integrated multisensory system is validated experimentally by the computation of deformable object models in which the measurements are merged. A formal dynamic model is discussed and a neural network representation model is presented. The results demonstrate the performance and suitability of the multisensory platform for the development of enhanced robotic hand capabilities when manipulating deformable objects.
An alternative minimum variance self-tuning controller is presented. The proposed controller, which is designed to combine a minimum variance strategy with a Dahlin controller, overcomes the shortcomings of the original version of minimum variance control and is found to be competitive with the generalized minimum variance controller. It tracks set-point changes with the desired speed of response, penalizes the excessive control action, and can be applied to non-minimum phase systems. The control algorithm is also extended to multi-variable systems. Robust pole-assignment can be incorporated to enable closed-loop system stability characteristics to be readily specified. Simulated results are presented to indicate the advantages of the proposed controller.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.