| In order to solving problem of the driving,balancing and trajectory tracking control of a single-ball mobile robot,the physical structure,kinematics and dynamics model are established firstly.Also,the control characteristics and influencing factors of the robot system are analyzed.Thus the scheme to improve the performance of the control system is determined based on the aspects of hardware selection and control algorithm.Secondly,the improvement and simulation analysis of self-balance,trajectory tracking control strategy and control algorithm are studied.Finally,the comprehensive experiments of single-ball mobile robot control is carried out base on the integration of the algorithms of self-balance and tracking.The main results are showed as fellow:Firstly,targeting the single-ball mobile robot developed by our academic team,the composition and structure of the system are introduced in detail,which is composed of high performance DSP,inertial detection unit HI219 embedded control chip and so on.The kinematic model between omnidirectional wheel and driving ball is established,and the dynamic model of the robot is deduced by using Eular-Lagrange equation.Also,the controllability,observability and stability of the robot system are analyzed by using Matlab.Then the control strategy of the system is determined.Secondly,on the basis of the kinematics and dynamics model of the robot established above,as well as the analysis of characteristics of the self-balancing control,the corresponding model is linearized,and the LQR controller optimized by genetic algorithm is designed by applying the LQR control method based on genetic algorithm optimization.The optimal global matrix Q and R parameters are obtained by the parallel global search ability of the genetic algorithm,which solves the problem that the weight matrix Q and R are difficult to determine in the LQR control method of the single-ball mobile robot.The simulation results show that: The performance of LQR controller optimized by genetic algorithm is better than traditional LQR controller in the attitude balance and position control of the robot,which indicates better real-time response and strong robustness.According to the under-driving and nonlinear characteristics of the robot trajectory tracking control,a high-gain observer is introduced to estimate the unmeasurable variables to improve the accuracy of the feedback variable while ensuring the asymptotic stability of the system.A high-gain observer is constructed.The underactuated sliding mode control method of separation theorem is designed and the corresponding trajectory tracking controller is designed.The simulation results show that the designed trajectory tracking controller realizes attitude balance and good trajectory tracking.At the same time,it proves that the controller has fast response and strong robustness from the adjustment time and steady-state performance indicators,and solves the system.The anti-interference ability is weak and the status is unmeasurable.Thirdly,the LQR controller based on genetic algorithm optimization and the underactuated sliding mode trajectory tracking controller based on the separation theorem of high gain observer are applied to the control system of single-ball mobile robot.The corresponding control software system was developed on the CCS platform,and the balance control experiment of the single-ball mobile robot was carried out.The results show that the attitude angle of the robot is basically within the range of ?5 degrees,and the steady-state adjustment time is kept within 3 s.Under the interference situation,the optimized LQR controller can realize the attitude angle control within ?10 degrees,and exceeds ?10 degrees,,the under-actuated sliding mode controller based on the high-gain observer separation theorem still guarantees the attitude balance adjustment.It can be seen that the single-ball mobile robot control strategy and control algorithm studied in this paper is feasible and has better stability,robustness and real-time. |
PDF Full Text Request