Study On Four Joint Biped Robot Position-posture And Automatic Obstacle Avoidance Control Algorithm

Intelligent robot is based on a new and multi-disciplinary technology including mechanics, electronics, detection and control science. Legged robot has strong qualities of flexibility and adaptivity to rugged terrain, which, however, are determined by robot's position-posture control and obstacle avoidance control.Although the legged robot's walking posture control was studied by some scholars and robot enthusiasts, most of the studies did not transcend the research of open-loop control. Similarly, the automatic obstacle avoidance control was investigated by many a scholar, the study mainly remains at the stage of roaming obstacle avoidance control with static single sensor. This dissertation was based on the study of biped robot with four joints, exporing closed loop position-posture control on the basis of examining plantar pressure triangles, gyroscope rolling angle and angle of pitch; also testing automatic obstacle avoidance control grounded on directional angle of gyro, GPS, ultrasonic and infrared ray. The following outlined the dissertation:1) The construction of biped robot with four joints was studied. The biped robot with four joints was constructed, whose parts include mechanical structure, main controller, pose detection system, obstacle detection system; the selected sensors(3D gyroscope, pressure sensor, ultransonic sensor, infrared distance measuring sensor) needed by the two systems mentioned obve were also closely examined.2) The close loop position posture control was researched. Firstly, based on the principle of both forward and inverse kinematics, A motion model for biped robot with four joints was constructed. Secondly, according to plantar pressure triangles, gyroscope rolling angle and angle of pitch, ZMP(zero movement point) were attained. Lastly, the stability of the biped robot with four joints was tested.3) Automatic obstacle avoidance control was examined. According to the theory of distance detection and the method of information fusion, a method of using both ultrasonic and infrared ray was propsoed in order to avoid the disadvantages when ultrasonic and infrared ray were used respectively. Regarding that distance detection sensor was not able to collect the edge information of the obstacles, security area was established on the basis of body repeated oscillation; then the aotomatic obstacle avoidance algorithm was worked out, based on security area and heading angle.4) The simulation experiment on biped robot with four joints and practical testing were carried out. Firstly, based on VS2012, a simulation system was designated, through which the close loop position posture control algorithm and automatic obstacle avoidance control algorithm were tested. Then the two control methods were tested by the biped robot with four joints. The conclusions suggested the control system proposed was valid.