The paper introduces an active approach for object recognition. Having control over an object, a robot is capable to reorient the object and acquire additional information, which would not be available without the active control of the object. By that the robot achieves higher recognition ratio. In the paper we additionally propose an innovative control of the robot arm/hand which results in optimal hand motion by exploitation of redundancy of the arm/hand and robotic vision.
COBISS.SI-ID: 22030375
The main result of this paper is an approach to learning parametrized sets of dynamic movement primitives based on a library of example movements. Learning was implemented by applying locally weighted regression where the goal of an action is used as a query point into the library of example movements. The proposed approach enables the generation of a wide range of movements that are adapted to the current configuration of the external world without requiring an expert to appropriately modify the underlying differential equations to account for perceptual feedback.
COBISS.SI-ID: 23181863
In this paper we were the first to combine the elements of motion imitation with dynamical primitives (DMPs) and adaptive frequency oscillators. The system for motion imitation extracts the frequency of the input demonstration signal and learns its waveform. The system is easy to use, numerically light and allows parametric modulation of the waveform, frequency and anchor point of the output signal. It allows adaptation to an external perturbation. The described system i is the first system based on dynamic primitives that does not need separate frequency extraction.
COBISS.SI-ID: 22774055
The publication describes the way how humans perform explosive movements such as vertical jumping and the transfer of the motion from a human to a robotic mechanism.
COBISS.SI-ID: 22098983
The paper describes the impreovement of the Wissler Human Thermal Model to enable simulation of the thermophysiological responses of clothed individuals conducting different types of activity in sub-zero environments. The author of the model, Prof. Eugene Wissler, assisted in introducing new algorithms in his basic model, which resulted in imnproved predictive power of the model. The model was tested with data collected during the evaluation of the Slovene Armed Forces Winter clothing ensemble at the Military training course in Pokljuka
COBISS.SI-ID: 22624551