earticle

영어

The human body has many degrees of freedom (1-3 degrees of freedom) per joint, enabling smooth and stable walking. However, unlike humans, robots require many more joints, and their design necessitates the control of numerous motors. However, controlling more motors leads to lower walking stability. Therefore, this paper studies a real-time control method for stable walking of a bipedal robot. To enable stable walking of a bipedal robot with 12 degrees of freedom, we study a method for generating and supplementing walking patterns between each joint using numerical analysis of the multi-joint robot's kinematics and learning-based AI. This research is to apply the adaptive control of neuron networks for the real-time attitude control of Multi-articulated robot. Multi-articulated robot is expressed with a complicated mathematical model on account of the mechanic, electric non-linearity which each articulation of mechanism has, and includes an unstable factor in time of attitude control. If such a complex expression is included in control operation, it leads to the disadvantage that operation time is lengthened. Thus, if the rapid change of the load or the disturbance is given, it is difficult to fulfill the control of desired performance. In this research we used the response property curve of the robot instead of the activation function of neural network algorithms, so the adaptive control system of neural networks constructed without the information of modeling can perform a real-time control. The proposed adaptive control algorithm generated control signs corresponding to the non-linearity of Multi-articulated robot, which could generate desired motion in real time.