PROJECT TITLE :

Toward Developing a Computational Model for Bipedal Push Recovery–A Brief

ABSTRACT:

The creature will negotiate with external push up to bound extent reactively. Grown up persons have higher push recovery capability than children and conjointly the professional wrestlers acquire higher push recovery capability than traditional soul. The acquired push recovery capability, therefore, is based on learning. However, the mechanism of learning is not known to us. Researchers around the planet are trying to explore this mystery through developing numerous models and implementing them on varied humanoid robots. All the models based on conventional mechanics and controls have inherent limitations. We have a tendency to believe acceptable computational model based mostly on learning can be able to effectively address this issue. Accordingly, we have collected extensively humanoid push recovery data using our innovative idea of exploiting the accelerometer sensor of sensible phone. Through our experiments, we tend to have studied the human push recovery by fusing knowledge at feature level using physics toolbar accelerometer of android interface kit. The subjects for the experiments were selected each as right handed and left handed. Pushes were induced from the behind with shut eyes to look at the motor action as well like open eyes to watch learning-primarily based reactive behaviors. A learning vector quantization-primarily based classifier has been developed to identify the coordination between numerous push and hip and knee joints.