Iros 2003
نویسندگان
چکیده
We present a self-valuing learning technique which is ca-pable of learning how to grasp unfamiliar objects andgeneralize the learned abilities. The learning system con-sists of two components which distinguish between localand global quality criteria for grasp points. The localcriteria are not object-specific while the global criteriacover physical properties of each object. In this case wepresent a generalization method of the learning param-eters based on a tree distance model for the medial axistransformations. The system is self-valuing, i.e. it ratesits actions by evaluating sensory information and the us-age of image processing techniques. An experimentalsetup consisting of a PUMA-260 manipulator equippedwith a hand-camera and a force/torque sensor, was usedto test this scheme. The system has shown the ability tograsp a wide range of objects and to apply pre-learnedknowledge to new objects.
منابع مشابه
Image-based reconstruction for view-independent human motion recognition
Introduction The problem of using computer vision to track and understand the behavior of human beings is a very important one. It has applications in the areas of human-computer interaction, user interface design, robot learning, and surveillance, among others.
متن کاملWalking navigation system of humanoid robot using stereo vision based floor recognition and path planning with multi-layered body image
To realize humanoid robots in unknown environment, sensor based navigation system is required as one of an essential function. This paper describes visionbased navigation system for humanoid robots, which has
متن کاملAbstracted biological principles applied with reduced actuation improve mobility of legged vehicles
ed Biological Principles Applied with Reduced Actuation Improve Mobility of Legged Vehicles Thomas J. Allen, Roger D. Quinn, Richard J. Bachmann, Roy E. Ritzmann Case Western Reserve University, Cleveland, Ohio, U.S.A. http://biorobots.cwru.edu, [email protected]
متن کاملThe first human-size humanoid that can fall over safely and stand-up again
This paper investigates a method through which a human-size humanoid robot can fall over backwards safely. Squatting-extending motion of legs reduce impact of falling and shock-absorbing parts of the robot keep the force at a permissible range. The robot could stand up itself again after falling.
متن کامل