柔性关节机械臂的非线性控制策略研究

针对外部干扰情况下的柔性关节机械臂非线性动力学模型,提出一种基于反演设计思想的递阶控制策略。把电机的输出角度向量作为关节子系统的控制变量,设计虚拟电机角度向量实现关节轨迹跟踪,同时在反演正定函数中综合积分项消除轨迹跟踪误差。计算实际的关节电机输出力矩,使电机输出角度跟踪虚拟控制量,通过设计自适应滑模变结构控制器消除系统不确定因素的影响。基于李雅普络夫稳定性理论证明系统的稳定性和轨迹跟踪误差的收敛性。     

柔性关节机械臂的非线性控制策略研究

针对外部干扰情况下的柔性关节机械臂非线性动力学模型,提出一种基于反演设计思想的递阶控制策略。把电机的输出角度向量作为关节子系统的控制变量,设计虚拟电机角度向量实现关节轨迹跟踪,同时在反演正定函数中综合积分项消除轨迹跟踪误差。计算实际的关节电机输出力矩,使电机输出角度跟踪虚拟控制量,通过设计自适应滑模变结构控制器消除系统不确定因素的影响。基于李雅普络夫稳定性理论证明系统的稳定性和轨迹跟踪误差的收敛性。     

高速开关阀驱动的气动肌肉关节的滑模变结构跟踪控制

建立了PWM高速开关阀控制的气动人工肌肉单关节的4阶SISO动态数学模型，针对实现轨迹跟踪目标，采用输入输出线性化方法得到相对阶数为3的等价系统，该等价系统零动态渐近稳定。由于系统具有参数不确定性和未建模动态特性，采用了带积分的滑模变结构控制。实验结果表明，采用基于等效降阶模型的鲁棒控制，当模型存在误差时，能在关节的整个运动范围内实现较高精度的轨迹跟踪，且对气源压力和负载变化等外加干扰具有良好的鲁棒性。     

一种并联机器人误差综合补偿方法

针对并联机器人轨迹规划和轨迹跟踪过程中,同时存在机构误差引起的期望轨迹与理想轨迹之间的偏差和非线性摩擦、负载变化等扰动因素引起的动态误差,提出一种并联机器人误差综合补偿方法:在轨迹规划过程中,基于并联机器人位姿误差模型将位姿误差补偿转化为驱动杆参数组合优化问题,进而利用粒子群算法寻优驱动杆参数,修正并联机器人期望轨迹;在轨迹跟踪过程中,设计基于自适应迭代学习控制算法的动态误差补偿策略,实现对期望轨迹的有效跟踪。在Stewart平台下基于ADAMS和Matlab进行仿真试验,在轨迹规划和轨迹跟踪过程中,分别修正期望轨迹偏差并补偿轨迹跟踪动态误差,实现并联机器人误差综合补偿。进一步,基于混联机床进行工件加工试验,验证方法对于提高并联机器人工作精度的有效性。     

变论域模糊补偿的机械臂积分终端滑模控制

为了减小机械臂在环境扰动、参数漂移和建模误差等影响下的轨迹跟踪误差,设计了基于积分终端滑模和变论域模糊补偿的组合控制器。采用拉格朗日方程建立了机械臂系统的动力学模型,制定了不确定因素影响下机械臂跟踪控制方案;设计的积分终端滑模控制器可以将系统初始状态限制在滑模面上,消除了控制过程的抖振并提高了跟踪速度;提出了自适应论域策略,该策略可以提高补偿力矩的输出细粒度,并将变论域模糊算法用于不确定因素补偿。经实验验证,变论域模糊补偿控制对关节1角位置的最大跟踪误差为0.103 rad,误差绝对均值为0.025 rad,对关节2角位置的最大跟踪误差为0.073 rad,误差绝对均值为0.012 rad,跟踪控制精度高于模糊补偿控制、RBF-BP控制和自适应鲁棒控制,验证了变论域模糊补偿控制方法的有效性和先进性。     

基于小脑模型神经网络的气动六自由度并联平台的复合控制方法研究

针对气动驱动系统的强非线性特性,本文采用基于小脑模型(CMAC)神经网络和PD的复合控制结构,用于气动六自由度并联平台各驱动关节的轨迹跟踪控制,以提高系统的跟踪精度和抗干扰能力。CMAC网络采用不均匀的输入量化函数,在零值误差附近增加量化等级,以提高系统的控制精度;而在误差较大处减小输入量化级数,从而在不增加存储空间的情况下,加大输入信号的范围。系统的仿真和实验结果都表明,与PD控制相比,该方法能有效地提高系统的轨迹跟踪精度。     

三关节移动机械臂轨迹跟踪控制的研究

三关节移动机械臂是由移动平台及固定在移动平台上的三关节机械臂组成的机器人系统。文章对三关节移动机械臂进行了运动学和动力学建模,提出了一种基于自适应动态滑模控制的轨迹跟踪控制方法。对受外界干扰影响的移动机械臂系统的数学模型进行了输入/输出线性化处理,并设计了一种带有不确定性上界估计的自适应动态滑模轨迹跟踪控制器。从理论上分析了闭环机器人控制系统的稳定性,并利用MATLAB轨迹跟踪控制仿真实验验证了所提出的轨迹跟踪控制方法的有效性。     

3自由度气动串联机械手的关节控制

3自由度气动机械手属关节串联式机器人,机械手在运动过程中,转动惯量、重力矩及关节间的耦合力矩等参数都会发生较大变化,影响了机械手末端的运动精度。针对这些问题,利用拉格朗日方程对机械手3关节进行动力学分析,得到多关节联动时单关节力矩方程。以腰部关节为例,通过对关节动力机构的数学模型线性化处理,采用状态反馈极点配置方法进行控制器设计,试验表明具有一定鲁棒性,但存在一定静态误差。分析产生误差的原因主要是干扰力矩的影响,根据单关节规划路径通过动力学模型得到补偿力矩,利用输入前馈对关节实施动态补偿。试验验证了方法的有效性,从结果可以看出,该组合控制策略抑制了扰动,提高了轨迹跟踪精度。     

基于气动伺服控制的虚拟现实运动模拟研究

提出了气动3-RPS并联平台应用于虚拟现实环境下运动模拟的解决方案。压缩空气和气动执行器作为一种廉价的驱动方式，但气动系统的强非线性、负载敏感、未知参数和死区等特性限制了运动模拟的精确度和逼真程度。从并联气动平台运动学动力学着手，基于光流图像处理技术生成关节空间的目标运动曲线，采用自适应鲁棒控制方法对气动并联运动平台进行实时轨迹跟踪的解决方案，相对传统算法，轨迹跟踪性能有了大幅提升。     

基于CMAC的气动人工肌肉变结构位置控制研究

气动人工肌肉是一种具有功率自重比/功率体积比大、响应快的新型气动元件,近年来已引起了人们广泛关注。然而,由于难于实现气动人工肌肉系统的精确控制,从而阻碍了其更加广泛的应用。在气动人工肌肉动态数学模型的基础上,提出采用基于CMAC的气动人工肌肉两层滑模变结构控制方法。CMAC神经网络用于学习气动人工肌肉系统的不确定信息,并作为前馈补偿使跟踪误差快速收敛,通过变结构控制消除网络的学习误差和不可重复随机干扰的影响,确保系统鲁棒性。试验结果表明了该方法的有效性和系统的鲁棒性。     

PARAMETRIC IDENTIFICATION OF PARALLEL MANIPULATOR WITH REDUNDANCY DRIVEN BY PNEUMATIC MUSCLES

A new parameter estimation algorithm is proposed for parametric identification of a parallel manipulator driven by pneumatic muscles with redundancy. Due to the special physical properties of the parallel manipulator studied, the regression model for parametric identification is characterized by multieollinearity, which will result in unreliable and inaccurate parameter estimations with large eovarianee if the conventional parameter estimation algorithm based on single error minimizing criterion is used. To improve the quality of parameter estimation and achieve high precise posture trajectory tracking control of the parallel manipulator, a new parameter estimation algorithm based on composite error minimizing criterion is developed in need of theoretical contractive forces of pneumatic muscles. The experimental results indicate that the proposed algorithm integrated with adaptive robust control could provide reliable parametric identification and greatly enhance the control accuracy in the trajectory tracking control of the parallel manipulator, and that the variation of known theoretical contractive forces of pneumatic muscles has slight influence on the control performance.     

Adaptive robust motion trajectory tracking control of pneumatic cylinders with LuGre model-based friction compensation

Friction compensation is particularly important for motion trajectory tracking control of pneumatic cylinders at low speed movement. However, most of the existing model-based friction compensation schemes use simple classical models, which are not enough to address applications with high-accuracy position requirements. Furthermore, the friction force in the cylinder is time-varying, and there exist rather severe unmodelled dynamics and unknown disturbances in the pneumatic system. To deal with these problems effectively, an adaptive robust controller with LuGre model-based dynamic friction compensation is constructed. The proposed controller employs on-line recursive least squares estimation（RLSE） to reduce the extent of parametric uncertainties, and utilizes the sliding mode control method to attenuate the effects of parameter estimation errors, unmodelled dynamics and disturbances. In addition, in order to realize LuGre model-based friction compensation, the modified dual-observer structure for estimating immeasurable friction internal state is developed. Therefore, a prescribed motion tracking transient performance and final tracking accuracy can be guaranteed. Since the system model uncertainties are unmatched, the recursive backstepping design technology is applied. In order to solve the conflicts between the sliding mode control design and the adaptive control design, the projection mapping is used to condition the RLSE algorithm so that the parameter estimates are kept within a known bounded convex set. Finally, the proposed controller is tested for tracking sinusoidal trajectories and smooth square trajectory under different loads and sudden disturbance. The testing results demonstrate that the achievable performance of the proposed controller is excellent and is much better than most other studies in literature. Especially when a 0.5 Hz sinusoidal trajectory is tracked, the maximum tracking error is 0.96 mm and the average tracking error is 0.45 mm. This paper constructs an adaptive robust controller     

Adaptive Robust Tracking Control of Pressure Trajectory Based on Kalman Filter

When adaptive robust control(ARC) strategy based on backstepping design is applied in pneumatic servo control, accurate pressure tracking in motion is especially necessary for both force and position trajectories tracking of rodless pneumatic cylinders, and therefore an adaptive robust pressure controller is developed in this paper to improve the tracking accuracy of pressure trajectory in the chamber when the pneumatic cylinder is moving. In the proposed adaptive robust pressure controller, off-line fitting of the orifice area and on-line parameter estimation of the flow coefficient are utilized to have improved model compensation, and meanwhile robust feedback and Kalman filter are used to have strong robustness against uncertain nonlinearities, parameter fluctuations and noise. Research results demonstrate that the adaptive robust pressure controller could not only track various pressure trajectories accurately even when the pneumatic cylinder is moving, but also obtain very smooth control input, which indicates the effectiveness of adaptive model compensation. Especially when a step pressure trajectory is tracked under the condition of the movement of a rodless pneumatic cylinder, maximum tracking error of ARC is 4.46 kPa and average tracking error is 0.99 kPa, and steady-state error of ARC could achieve 0.84 kPa, which is very close to the measurement accuracy of pressure transducer.     

Fuzzy Torque Control of the Bionic Flexible Manipulator Actuated by Pneumatic Muscle Actuators

A bionic flexible manipulator driven by pneumatic muscle actuator(PMA) can better reflect the flexibility of the mechanism.Current research on PMA mainly focuses on the modeling and control strategy of the pneumatic manipulator system.Compared with traditional electro-hydraulic actuators,the structure of PMA is simple but possesses strong nonlinearity and flexibility,which leads to the difficulty in improving the control accuracy.In this paper,the configuration design of a bionic flexible manipulator is performed by human physiological map,the kinematic model of the mechanism is established,and the dynamics is analyzed by Lagrange method.A fuzzy torque control algorithm is designed based on the computed torque method,where the fuzzy control theory is applied.The hardware experimental system is established.Through the co-simulation contrast test on MATLAB and ADAMS,it is found that the fuzzy torque control algorithm has better tracking performance and higher tracking accuracy than the computed torque method,and is applied to the entity control test.The experimental results show that the fuzzy torque algorithm can better control the trajectory tracking movement of the bionic flexible manipulator.This research proposes a fuzzy torque control algorithm which can compensate the error more effectively,and possesses the preferred trajectory tracking performance.     

气动机械手轨迹的Terminal滑模控制方法

应用Terminal滑模控制方法对三轴直角坐标型气动机械手进行连续轨迹控制。首先建立了气动位置伺服系统的数学模型,然后运用Terminal滑模控制对机械手进行轨迹控制。仿真研究结果表明,采用高阶非线性的Terminal滑模控制方法,可以使该机械手对空间直线轨迹的跟踪误差只在未达到收敛点的时间段内较大,在到达收敛点后能完全跟踪目标轨迹。     

空间刚柔耦合并联机构的逆动力学建模与控制

为了提高3-RRRU空间刚柔耦合并联机构的轨迹跟踪精度,提出了一种基于瞬态刚体校正法的逆动力学模型求解方法来构建该机构的非线性控制策略。首先,利用自然坐标法和绝对节点坐标法建立该机构的非线性逆动力学模型,它考虑了各支链柔性空间梁单元的剪切效应,并能描述柔性梁的大范围非线性弹性变形。然后,通过分析刚柔耦合动力学模型在求解过程中出现的相容性问题,结合自然坐标法与理想运动学模型,提出了瞬态刚体校正法并求出逆动力学模型的稳定数值因果解。最后,基于该数值解构建并联机构的非线性控制策略,通过仿真与实验验证了该方法的可行性与有效性。仿真与实验结果表明:逆动力学方程组的求解精度为10-6,约束方程的相容误差为10-8;与刚性并联机构的控制方法相比,该方法在圆形轨迹下的最大跟踪误差降低了0.465mm,圆度误差降低了0.416mm。结果表明:该求解方法解决了闭链机构多体动力学方程的违约问题,有效地改善了系统的综合收敛性能,所构建的控制策略提高了并联机构的轨迹跟踪精度。     

气动三自由度并联平移型机器人的同步控制

气动三自由度并联平移机器人由固定平台、动平台、3个辅助臂和3个驱动臂组成,3个辅助臂约束机器人动平台的转动,3个气缸驱动机器人动平台沿x, y, z方向的运动。根据并联平移机器人的机械结构及运动特征,得到动平台的运动学正解和逆解。在并联平移机器人的运动控制中,不仅要减少单个驱动臂的跟踪误差,还需考虑驱动臂间的运动协调性。由于气体的可压缩性、负载的时变性和摩擦力等因素,精确的驱动臂模型难以建立,因此针对单个驱动臂设计了线性自抗扰控制器;此外,根据机器人结构特点设计基于驱动臂跟踪误差分解组合的协调控制器,解决了运动协调的问题,进一步减小了空间轨迹的跟踪误差。实验证明了该控制策略的有效性。