Study On The Two Wheel Robot With Center Of Gravity Self-adjusting Mechanism

Two wheeled self-balancing robot is a kind of wheeled robot based on inverted pendulum structure.The structural characteristic of coaxial parallel arranged driven wheels endows the robot with uncomplicated and compact composition and flexible and superior mobility performance.The merits above also allow the robot to be regarded as an ideal mobile robot platform with comparative practical value and wide application prospects.Simultaneously,as the two wheeled robot system is a complicated nonlinear under-actuated system,it widely attracts researchers around the world studying in intelligent algorithm research,mobility control study and relative fields.With the rapid development of robot system,sense technology,intelligent algorithm,data service and other robot techniques,two wheeled self-balancing robot still remains tremendous research value and study significance.In this paper,we proposed a center of gravity self-adjusting two wheel robot with fundamental characteristics of two wheel self-balancing robot.In addition,it employs the center of gravity self-adjusting mechanism and advanced intelligent control strategy to further improve its mobility performance and practicability as well as the fault detection technic and redundant control system to improve the stability of system operation.Main contents and results achieved in this paper are as follows:1)On the basis of mobility characteristics and technical requirements of the two wheel self-balancing robot,we proposed a center of gravity self-adjusting two wheel self-balancing robot mobile platform design scheme to further improve the mobility and practical application value of the mobile robot platform.The robot system construction method is introduced in details together with the specific technical realization method.2)In view of analysis and study of robot system architecture,a two level hierarchical system structure is employed according to ROS operation system to satisfy the robot function demands of increasingly robot complication and intelligence.As the advantages of the distributed framework of ROS operation system,it is profitable to increase the robot system operational efficiency.Furthermore,the modulation design concept can improve thesystem extensibility.3)We built the kinematics and dynamics models on the basis of analysis and discussion of the proposed robot system kinematics and dynamics issues.The Lagrange equation is adopted to build the dynamics model on the condition of different motion states and road conditions.Then,the achieved dynamic model is simplified and decoupled in accordance with motion feature.4)The Mamdami-like fuzzy control method is employed to design the robot balance controller,velocity controller and turning controller respectively in order to achieve better motion control performance on the basis of center of gravity self-adjusting two wheel robot dynamic model.Simultaneously,corresponding traditional PID controllers are also designed for comparisons.The system simulation and analysis realized on Matlab platform to testify the vitality of the proposed control strategy and the superiority of the Mamdami-like fuzzy controller performance.5)Because of the instability of the inverted pendulum mechanism of two wheel self-balancing robot,a data based improved version of principal component analysis method is proposed for the robot fault detection according to the discussion and analysis of fault cause,fault position and fault phenomena.The redundant control system as an assistant system raises the system stability and security consequently.The simulation results testify the correctness and availability of the designed fault detection method.6)The constructed physical experiment system provides the platform to implement the specific experimental study.The experiment results prove the success of designed robot performance comprehensively.Moreover,it also proves the effectiveness and rationality of the theory analysis and technical method illustrated in this paper.