The petri nets and markov chains approach for the walking robots dynamical stability control

This paper investigate Zero Movement Point (ZPM) dynamic control of walking robots and develop an open architecture of real time control multiprocessor systems, in view of obtaining new capabilities for walking robots. Considering the complexity of the movement mechanism of a walking robot being a repetitive tilting process with numerous instable movements can lead to its turnover on an uneven terrain. The control system architecture for the dynamic robot walking is presented in correlation with a stochastic model of assessing system probability of unidirectional or bidirectional transition states, applying the non-homogeneous/non-stationary Markov chains. The capability of the time-dependent method for describing a multi-state system and assessing the operational situation of a robotic system is evaluated based on a case study. The rationality and validity of the proposed model are demonstrated via an example of quantitative assessment of states probabilities of an autonomous robot. The results show that the proposed new strategy of the walking robot control systems for slope movement and walking by overtaking or going around obstacles has increased the robot´s mobility and stability in real.