The paper deals with a system for simultaneous people tracking and posture recognition in cluttered environments in the context of human-robot interaction. We adopt no particular assumptions on the movement of a person nor on its appearance, making the system suitable to several real-world applications. The system can be roughly subdivided into two highly correlated phases: tracking and recognition. The tracking is concerned with establishing coherent relations of the same subject between frames. We adopted a version of particle filters known as Condensation algorithm due to its robustness in highly cluttered environments. The recognition phase adopts a modified eigenspace technique in order to classify between several different postures. The system has demonstrated to be robust in different working conditions and permits to exploit a rich interaction between a human user and a robotic system by means of a set of natural gestures.

The evolution of the crystallization front from a planar to a dendritic one as a function of the GL/(Rc0) parameter was investigated during the crystallization of Al-Cu alloys by the vertical Bridgman method. Six series of alloys with different initial composi- tions of Cu were solidified at different growth rates. A mathematical model for the heat transfer during vertical Bridgmen crystal growth was developed. The model was solved using the finite element method. The temperature gradient in the melt at the beginning of crystal growth was calculated using the obtained model. Discrete stages of the crystalliza- tion front were identified in the experiments, as the ratio GL/(Rc0) decreased.