摘要
在机器人无侧向移动的条件下,以左右轮的驱动力为输入量,建立两轮足球机器人平动和转动的离散时间状态空间模型;将免疫算法与预测控制结合,提出了一种免疫预测控制方法,在所建立的状态空间模型的基础上,对机器人进行平动和转动控制。该算法利用克隆选择算法实现预测控制的滚动优化,避免了带约束的目标函数最优解及逆矩阵的复杂推导。仿真结果表明,该方法对机器人平动和移动控制能达到理想的控制效果,对外部干扰及建模误差具有很好的适应性。
Under the condition that the robot has no side movements,using the driving forces of the left and right wheels as the input parameters,the translatory and rotary discrete time state space models of a 2-wheel soccer robot are established.With the combination of immune algorithm and predictive control,an immune predictive control strategy is proposed to control the robot＇s translation and rotation on the basis of the state space model.The immune predictive control strategy utilizes a clonal selective algorithm to implement the rolling optimization of the predictive control,which can avoid the complicated derivation of inverse matrix and derivation of optimum solutions of constrained objective functions.Simulation results show that the proposed method can achieve good performance in the control of the robot＇s translation and rotation,as well as a good adaptability to external disturbances and modeling error.
出处
《南昌大学学报（理科版）》
CAS
北大核心
2010年第4期389-392,共4页
Journal of Nanchang University(Natural Science)
基金
广东省自然科学基金资助项目(07008360)
广西工学院博士基金资助项目(200605)
