【作者】 龚文涛；

【导师】 张铁；

【作者基本信息】 华南理工大学 ， 机械设计及理论， 2017， 硕士

【摘要】 机器人已广泛应用于现代工业,随着机器人和自动化技术的发展,其高速运动时的准确性和稳定性日益成为检验机器人性能的重要指标,是机器人的运动控制系统必须考虑的因素。在高速运动的情况下,机器人的非线性控制特性十分明显,并且实际存在的各种复杂不确性因素也严重影响着机器人的控制精度,因此对工业机器人进行运动优化及基于动力学的控制方法研究,以实现高速高精度的控制目标具有重要的理论意义和实用价值。根据工业机器人运动控制的实际特点,在运动学和动力学两方面对其高速高精度控制的基础理论展开研究,分析了连续运动路径的前瞻速度控制方法和直线、圆弧等运动路径的笛卡尔空间插补方法。对于直线和圆弧等运动路径,引入了S型柔性加减速规划方法和多次样条曲线速度规划方法,以减小机器人启停时受到的冲击和振动;针对NURBS(非均匀有理B样条)曲线,研究了基于曲线几何特性的前瞻分段速度规划方法。最后研究了基于牛顿-欧拉迭代方程建模的机器人动力学参数辨识方法,建立了基于动力学模型的前馈力矩控制模型。研究的主要内容如下:(1)研究了连续路径的前瞻速度控制方法,在各相邻路径之间添加过渡曲线,并规划出各段路径起始和终点处速度。同时也研究了直线、圆弧、过渡样条曲线等运动路径的笛卡尔空间实时插补方法。仿真结果表明,该方法避免了机器人在路径过渡处的频繁启停,提高了机器人运行的速度。(2)对于直线和圆弧路径,研究了S型柔性加减速速度规划方法,以减小机器人启停时受到的刚性冲击;而为了应对机器人带大负载高速运动的场合,研究了多次样条曲线速度规划方法;为了得到无冲击且柔性很好的加减速效果,研究了三角函数速度规划方法;对于NURBS曲线,采用全局插值和一阶泰勒展开的方法生成NURBS曲线,然后研究了基于曲线向心加速度和弓高误差限制等几何特性的前瞻分段速度规划方法。实验结果表明,以上速度规划方法,在一定程度上减小了机器人启停时受到的冲击,提高了工业机器人的位置控制精度。(3)研究了基于牛顿-欧拉迭代方程建模的机器人动力学参数辨识方法,通过数值解法得到带基础动力学参数的动力学方程,解非线性最优化问题得到最优傅里叶激励轨迹,使用带权最小二乘法解超静定方程得到机器人基础动力学参数。建立基于机器人动力学模型的前馈力矩补偿模型,根据机器人的理论轨迹,建立缓存队列,根据中心差分法计算各关节的角速度和角加速度量。在伺服电机位置控制模式下,在每个插补周期实时计算机器人各关节的前馈补偿力矩,并发送到电机底层的电流环,以作为电机反馈力矩的补偿。实验结果表明,基于动力学模型的前馈力矩控制可以在很大程度上提高机器人的响应速度,减小机器人各关节的轨迹跟踪误差和机器人停止时各关节的抖动。(4)搭建了基于x86平台和嵌入式实时操作系统的机器人运动控制系统平台,分别对前述高速高精度运动控制算法进行了相关实验验证。实验结果验证了上述算法和理论的可行性和有效性。更多还原

【Abstract】 Robot has been widely used in modern industry.With the development of robot and automation technology,the accuracy and stability of high-speed motion are becoming an important index of robot performance,and it is a factor that should be considered in the motion control system of a robot.In the case of high speed motion,the nonlinear control characteristics of the robot are very significant,and the actual complexity of the complex factors also seriously affect the control accuracy of the robot,it is of great theoretical significance and practical value to research on motion optimization and dynamic control method of industrial robot to meet the high-speed high-precision control objectives.In the light of the practical characteristics of motion control of industrial robot,the basic theory of high-speed and high-precision control is researched in both kinematics and dynamics.At first,research the forward velocity control method of continuous motion path and the Cartesian space interpolation method of straight line and Arc motion path.And then,introduce S-type flexible acceleration/deceleration planning method and multiple spline curve velocity planning method to reduce the impact and vibration of robot when it starts and stops.As for the NURBS curve(non-uniform rational B-spline),a forward-speed segmentation method based on the geometric properties of the curve is studied.Finally,a kinematic parameter identification method based on Newton-Euler iterative equation is studied,and a feedforward torque control model based on dynamic model is established.The main contents of the study are as follows:(1)The forward velocity control method of continuous path is researched,and the transition curve is added to each adjacent path,and the velocity of the start and end points of each path is planned.And the Cartesian space real-time interpolation method of motion path such as straight line,Arc and transition spline curve is also researched.The simulation results show that the method avoids the frequent start and stop of the robot at the transition of the path and improves the velocity of the robot.(2)For the straight and circular path,the S-type flexible acceleration/deceleration planning method is researched to reduce the rigidity impact of the robot when it starts and stops,and in order to cope with the high-load and high-speed movement of the robot,multiple spline curve velocity planning method is proposed.In order to obtain the effect of acceleration and deceleration without impact and flexibility,the trigonometric function velocity planning method is researched.For NURBS curves,the curves are generated using global interpolation and first-order Taylor expansion.And then,the forward-speed segmentation method based on geometric characteristics such as centripetal acceleration and chord error is established.The experimental results show that the above-mentioned velocity planning method can reduce the impact of start and stop period of the robot and improve the position control precision of the industrial robot to a certain extent.(3)The dynamic parameters of the robot based on the Newton-Euler iterative equation are researched and the kinetic equation of the basic dynamic parameters is obtained by numerical solution.Then solve the nonlinear optimization problem to obtain the optimal Fourier excitation trajectory,use the weighted least squares method to solve the stationary equation to obtain the basic dynamics parameters of the robot.The feedforward torque compensation method based on the robot dynamics model is established,according to the theoretical trajectory of the robot,the dynamic buffer queue is established,and the angular velocity and angular acceleration of each joint are calculated according to the central difference method.In the servo motor position control mode,the feedforward compensation torque of each joint of the robot is calculated real time in each interpolation cycle and sent to the current loop at the bottom of the motor to compensate for the feedback torque of the motor.The experimental results show that the feedforward torque control based on the dynamic model can improve the response speed of the robot to a large extent,and reduce the tracking error of the joints of the robot and the jitter of each joint when the robot is stopped.(4)Built a robot control system based on the x86 platform and embedded real-time operating system and in this platform,the high-speed and high-precision motion control algorithm is experimentally verified.The experimental results prove the feasibility and effectiveness of the above algorithms and theories.更多还原