【作者】 陈磊；

【导师】 柳向斌；

【作者基本信息】 北京交通大学 ， 控制科学与工程， 2024， 硕士

【摘要】 机械臂被广泛应用于工业制造、家庭协作、医疗以及军事等领域。由于机械臂本身非线性、强耦合等特点为其控制器设计带来了诸多挑战。本文针对由机械臂模型参数化不确定、外界扰动以及交互环境位置不确定等所引起的机械臂位置跟踪精度与力跟踪精度问题,从机械臂正向以及逆向运动学建模、动力学建模、轨迹规划、位置控制和柔顺控制等关键技术展开研究,最终实现了机械臂在自由空间下对期望轨迹的准确跟踪,在接触空间下对期望恒力的力跟踪。本论文的主要工作如下:首先,针对六自由度机械臂,利用标准DH参数法及齐次变换矩阵建立了机械臂的正向以及逆向运动学模型,推导出了逆向运动学模型的解析解;在运动学模型基础上,利用Lagrange方法建立了机械臂的动力学模型。其次,在机械臂关节空间中,本文针对机械臂末端执行器点到点的任务进行了梯形轨迹规划、三次多项式轨迹规划、五次多项式轨迹规划以及贝塞尔样条曲线轨迹规划,并比较了不同算法的优缺点;在笛卡尔空间中实现了对直线以及圆弧基本路径的轨迹规划,并利用五次多项式轨迹规划方法对机械臂末端执行器轨迹进行了二次优化,避免了机械臂启停时力矩过大造成对关节造成冲击。再次,针对带有参数化不确定以及受到外界扰动的机械臂系统,本文设计了一种带有输出受限的鲁棒自适应控制策略以实现准确的轨迹跟踪控制并对关节位置的输出约束,算法结合不连续投影,对自适应更新律进行了修正以保证投影后的参数估计值不超过所设定的参数范围。最后在MATLAB中开展了仿真实验,以验证算法的有效性。最后,为实现机械臂的柔顺控制,本文引入基于位置的阻抗控制作为基本控制框架。将带有输出约束鲁棒自适应控制器作为内环控制器,保证位置跟踪精度;将导纳控制器作为外环控制器,并引入了自适应项使阻抗控制器主动适应接触面位置变化,保证力跟踪控制的鲁棒性。最后通过MATLAB仿真实验表明所提出的算法具有较好的力跟踪性能和环境适应性。图61个,表4个,参考文献89篇更多还原

【Abstract】 Robotic manipulator has been widely used in industrial manufacturing,home collaboration,medical and military fields.However,the nonlinear and strong coupling characteristics of robotic manipulators bring many challenges for controller design.Considering the robotic manipulator position tracking accuracy problem and force tracking accuracy problem caused by model uncertainties,external disturbance,and the uncertainty of the position of the interaction environment,this thesis carries out study on the key technologies which include forward and inverse kinematic modeling,dynamics modeling,trajectory planning,trajectory tracking,and compliance control.Ultimately,this thesis achieve accurate tracking of the desired trajectory in free space and accurate tracking of the desired constant force in contact space.The main work of this thesis is as follows:Firstly,for the six-degree-of-freedom robotic manipulator,the forward and inverse kinematic modes of the robotic manipulator are established by using the standard DH parameter method and the transform matrix.And the analytical solution of the inverse kinematic model is obtained;based on the kinematic model,the dynamics model of the robotic manipulator is established by using Lagrange method.Secondly,this thesis carries out trapezoidal trajectory planning,cubic polynomial trajectory planning,quintic polynomial trajectory planning and B-spline trajectory planning for the point-to-point task of the robotic manipulator end-effector.Furthermore,this thesis completes the trajectory planning for the straight line and the circular curve in Cartesian space,and use the quintic polynomial trajectory planning method to secondly optimize the trajectory of the robotic manipulator end-effector to avoid excessive torque that may damage the joints when the robotic manipulators start and stop.Thirdly,for the robotic manipulator system with parameterization uncertainty and external disturbance,a robust adaptive control strategy with output constraints is designed to achieve accurate trajectory tracking control and output constraints on the joint position.In addition,combined with discontinuous projection,the adaptive updating law is modified to ensure that the parameter estimation does not violate the set parameter ranges.Simulation experiments are performed in MATLAB to verify the effectiveness of the control strategy.Finally,in order to achieve flexible control of the robotic manipulator,this thesis introduces position-based impedance control as the fundamental control framework.The robust adaptive controller with output constraints is used as the inner loop to guarantee position tracking accuracy.The admittance controller is used as the outer loop,and an adaptive term is introduced to enable the impedance controller to adapt to changes in the contact surface actively to ensure the robustness of the force tracking control.The results of MATLAB simulation experiments demonstrate that the proposed algorithm exhibits excellent force tracking performance and environmental adaptability.There are 61 figures,4 tables and 89 references in this thesis.更多还原