Research On Stable Gait Planning Of Bipedal Robot And The Self-localization Of Multi-robots In Indoor Environment

With the rapid development of science and technology as well as the coming of aging society, all kinds of service robots with unique characteristics are appearing into our daily life. Among all the robots, the biped walking robot with human resembles and movement characteristics stands out the most. It is not difficult to foresee in the near future that humanoid robot will act like an independent "person", and coexist in harmony with human beings. However, austere challenges go hands in hands with beautiful longings, and the artificial intelligence era is still far from coming up. In this article, to meet the requirements of letting the humanoid robot to walk stably and naturally, and locating itself in indoor environment efficiently, studies on kinematics modeling, dynamics modeling, the gait planning of the omnidirectional biped walking and the self-location in indoor environment are carried out.According to the structure characteristics of the biped robot Nao and its walking manners, we build the kinematics and dynamics model of the lower limbs of Nao robot. Specific to the six freedom joint configuration of the lower limb’s single leg, this paper proposes a method which can converted a non-orthogonal configuration with three joints into a standard orthogonal configuration, thus to simplify the process of finding the closed inverse kinematics solution. For similar problems, this method provides a general solving process. Based on the ZMP criteria and linear inverted pendulum(LIPM) model, we implement research on the changing rule of the pressure center during human walking, we proposed a next step planning method, which can generate real-time online gait but reduce the calculations on a large scale. Through the simulation experiments and physical prototype experiment, the validity of the bipedal gait planning algorithm is verified.According to the basic sensors equipment of Nao, we design a self-location algorithm for the robot Nao to locate itself. Only with the using of the included 2D camera and the artificial landmarks, we realize the self-location with an acceptable accuracy without any external auxiliary equipment. With the using of multi-threads communication technology, we realize the multi-robot communication between each robots. Based on single-robot position results and the multi-robot communication technology, we realize the multi-robot formation control and the movement coordination. Finally through the formation and navigation experiment of multi-robot, the self-localization algorithm and multiple robot coordination mechanism is verified.