To be effective, action learning should not be limited to direct replication of movements obtained during training but must also enable the generation of actions in situations a robot has never encountered before. This paper describes a methodology that enables the generalization of the available sensorimotor knowledge using statistical techniques.
COBISS.SI-ID: 23918375
In this article, we presented a compete concept of learning action primitives to recognize and synthesize actions. One of the main novelties in this work is the detection of primitives in a unified framework, which takes into account objects and actions being applied to them. We proposed an unsupervised learning approach for action primitives that automatically identifies the observed primitives and the associated action grammar.
COBISS.SI-ID: 23725095
This article's main contribution is a novel dynamic model that considers friction effects. The presented model includes all three working principles of the device: free rotor mode and both modes of rotor rolling in the housing. The work introduces models with one and two degrees of freedom actuation, both of which are suitable for laboratory control experiments. An estimation of the friction is discussed, and both the simulation and the experimental results are presented to evaluate the models.
COBISS.SI-ID: 24178983
The manuscript describes the effect of anti-histaminic and anti-cholinergic substances on heat production and heat loss in humans. On the basis of these results, the manuscript enhances our knowledge of the neural network governing human temperature regulation, published earlier (Mekjavic &Eiken, Journal of Applied Physiology, 100: 2065-2072).
COBISS.SI-ID: 23693607
The manuscript evaluates a novel approach for determining the optimal insulation of protective clothing. The mathematical model accounts for the autonomic thermoregulatory responses, which it determined on the basis of algorithms derived on the basis on human experiments.
COBISS.SI-ID: 23969831