基于动力学的拟人肩关节动载协调分配优化研究

针对并联机构动载分配优化存在的问题,对一种拟人肩关节机构的运动学、动力学、机构性能等方面进行了研究,提出了一种动载协调分配优化方法。首先结合拉格朗日方程和虚功原理建立了动力学模型,并利用动力学仿真软件验证了动力学模型的正确性;然后基于动力学模型,建立了肩关节机构的动力学性能指标和力映射性能指标,并应用加权求和法将其转换成综合性能指标;最后采用Dijkstra算法得到了性能最优的轨迹,并综合考虑时间、能耗和力矩波动,采用遗传算法优化了该机构的广义时间,从而确定了各个关节的力矩、角位移和角速度。研究结果表明:该优化方法能够实现拟人肩关节在时间最短、能耗最小、性能最优的条件下完成运动,且该优化方法也适用于其他并联机构。     

飞行器折叠翼机构展开性能的优化及实验

为了优化飞行器折叠翼机构的展开性能,对飞行器的折叠翼机构进行分析,并提出了优化其展开性能的方法。建立了折叠翼机构展开的理论模型和动力学仿真模型,对折叠翼机构展开过程进行了动力学分析,并进行了实例计算。分析了影响折叠翼机构展开性能的因素,用正交试验的方法对折叠翼机构进行了优化设计,得到了最优折叠翼机构方案,并对最优方案进行了仿真分析。最后,按照最优方案折叠翼机构进行了改进,实验测试了折叠翼机构的展开性能。实验测得折叠翼机构展开时间为0.128s,测点位置应力分别为82 MPa和92 MPa;动力学仿真得到折叠翼机构展开时间为0.122s,相应测点位置应力分别为85MPa和96MPa,两项测试误差在5%以内。得到的结果表明折叠翼机构满足机翼展开稳定、快速,应力和冲击力小的要求,为飞行器性能的调高奠定了基础。     

■高速并联机器人动力尺度综合

研究一种新型四自由度高速并联机器人动力尺度综合方法。该机构以■为支链且具有单动平台结构,与采用双动平台结构的此类机构相比,末端动平台质量更轻,从而具有良好的加减速性能。建立机构运动学和动力学模型,并构造出可反映机构运动和力传递特性的链内/链间压力角,以及以单轴最大驱动扭矩全域最大值最小的动力学性能评价指标。在此基础上,综合考虑装配尺寸等几何约束和运动学性能约束,建立了以使机构动力学性能最优的机构尺度综合模型。在探究压力角和动力学性能随尺度变化规律后,给定具体约束条件,通过遗传算法优化得到一组最优尺度,并基于在Lamé抓放轨迹上运动仿真完成伺服电动机参数预估。     

■高速并联机器人动力尺度综合

研究一种新型四自由度高速并联机器人动力尺度综合方法。该机构以■为支链且具有单动平台结构,与采用双动平台结构的此类机构相比,末端动平台质量更轻,从而具有良好的加减速性能。建立机构运动学和动力学模型,并构造出可反映机构运动和力传递特性的链内/链间压力角,以及以单轴最大驱动扭矩全域最大值最小的动力学性能评价指标。在此基础上,综合考虑装配尺寸等几何约束和运动学性能约束,建立了以使机构动力学性能最优的机构尺度综合模型。在探究压力角和动力学性能随尺度变化规律后,给定具体约束条件,通过遗传算法优化得到一组最优尺度,并基于在Lamé抓放轨迹上运动仿真完成伺服电动机参数预估。     

基于运动学、刚度和动力学性能的并联机构有序递进三级优化设计及其应用

提出一种基于运动学、刚度和动力学性能的并联机构有序递进三级优化策略,即分别应用遗传算法（GA）、粒子群算法（PSO）、差分进化算法（DE）三种智能算法,以工作空间性能和运动/力传递性能为目标的尺度参数优化方法,又在优化尺度参数基础上,以机构承载刚度和整体刚度为目标进行构件截面参数优化,再在优化尺度参数和截面参数基础上,以动力学灵巧度和能量传递效率为目标进行构件质量参数优化,从而使得机构的尺度、截面及质量参数达到最优。以一种零耦合度三平移一转动（SCARA）并联操作手为例,运用矢量法建立了并联操作手的运动学模型,通过工作空间性能和运动/力传递性能两个运动学指标,优化其尺度参数;运用虚拟弹簧法得到该并联操作手的刚度模型,通过对三维模型的参数识别得到其柔度矩阵,并分析了承载刚度、整体刚度两个刚度性能指标,优化了构件的截面参数;利用动力学普遍方程导出了该并联操作手的动力学模型,通过动力学灵巧度和能量传递效率两个动力学性能指标,优化了构件的质量参数。最终,该并联操作手的运动学、刚度和动力学综合性能达到最优,也得到了一些尺度、截面及质量参数的设计准则。     

铁道车辆动力学模型设计及优化分析

采用多体动力学软件SIMPACK建立铁道车辆动力学模型,并对其进行优化分析,调用SIMPACK计算不同车辆悬挂参数下的车辆动力学性能,包括安全性、稳定性、舒适性等指标。采用脱轨系数、倾覆系数、轮轨横向力评价指标对车辆运行安全性进行评定,优化分析采用最优试验设计方法构建了近似模型,经过优化计算,车辆的动力学综合性能提高了30%以上,每一个动力学指标都有不同程度的提高。     

混合机构式机器人的动力学建模及参数优化

建立了一种混合机构式机器人的动力学模型 ,并研究了其运动学性能。基于动力学参数对其运行速度有较大的影响 ,采用改进遗传算法对动力学参数进行了优化 ,得到了动力学参数的最优解 ,为机构设计和系统控制提供了理论依据。     

三自由度拟人并联机构的尺度参数优化研究

为提升并联机构的运动学与静力学综合性能,提出了一种基于空间模型技术的并联机构多目标尺度参数优化方法。首先,根据人机工程学设计了一种2-PUS/RS三自由度并联机构,并基于螺旋理论分析了机构2种类型支链的自由度,计算了并联机构的自由度;采用欧拉角描述了动平台姿态,推导了机构的位置反解;在肘、腕关节的转动副和驱动器约束条件下,利用搜索法得到了机构的工作空间;然后,定义了角速度传递、角速度各向同性性能评价指标,绘制了评价指标在工作空间中的等高线分布图;最后,利用空间模型技术无量纲化尺度参数,绘制了各项全域性能图谱,并采用加权求和法对其进行了多目标参数优化,得到了机构的最优结构参数。研究结果表明:优化后机构的各项指标大幅提升,全域角速度传递性能、力矩传递性能和各向同性性能分别提高了16%、80%、40%,验证了该尺度参数优化方法的有效性。     

一种新型4自由度高速并联机器人设计方法

提出一种可实现SCARA运动的新型4自由度高速并联机器人的动平台,采用双螺旋齿轮传动原理,将两相对支链驱动转化为末端的自转运动。针对该机器人的动力尺度综合问题,在建立运动学和刚体动力学模型基础上,利用直接和间接雅可比矩阵的行列式,构造出两类揭示机构奇异状态并定量描述机构链内与链间的力传递特性的空间压力角。基于可反映系统动力学特性的性能评价指标,在几何约束和力传递性能约束条件下,借助多目标优化的方法获得可保证机构运动学和动力学性能的最优尺度参数。     

Delta机器人的尺度综合方法研究

尺度综合是并联机器人运动学设计的最终目标.为此,本文针对Delta机器人提出了一种最优尺度综合方法.首先对机构进行运动学分析得到局部灵活度的性能评价指标,其次将局部性能评价指标综合为全域性能指标,并将尺度综合问题归结为一类参数优化问题.上述方法对这类高速轻型并联机器人以及其它并联机构的运动学设计理论有一定指导意义.     

基于能耗指标的拟人机器人步态优化与分析

通过引入基于平均功率、平均功率偏差、平均力矩损耗三个能耗指标对拟人机器人步态进行了优化与分析.基于拉格朗日方程推导了各个关节驱动力矩的计算公式,建立了拟人机器人步态优化的数学模型.使用Matlab作为优化与仿真工具,得到了一組拟人机器人步态参数最优解,分析了拟人机器人各个步态参数与拟人机器人能量消耗之间的关系.并给出了拟人机器人步态优化与分析的数值实例.     

冗余仿人臂避关节物理约束的一种逆运动学问题求解方法

冗余仿人臂有无穷多组逆运动学解满足末端位姿要求,但由于其连杆机构与形状的设计仿人臂每个自由度都有关节位置物理约束。为使所求取的逆运动学解最大化地远离仿人臂关节限位,提出冗余仿人臂(8-DOF)一种几何-数值迭代相结合求解逆运动学问题方法,该方法在计算出仿人臂逆运动学解几何表达式基础上,以仿人臂的“远离限位度”指标构建适应度函数并以此为寻优目标,通过粒子群优化方法(Particle swarm optimization,PSO)搜索冗余关节角最优组合,并获得满足仿人臂末端位姿要求的最优逆运动学解。该方法不仅解决了冗余仿人臂逆运动学多解问题,而且所求关节角不存在理论误差、求解速度快及所求关节角远离关节位置限位裕量大的优点,仿真试验验证了该方法有效性。     

Posture optimization for a humanoid robot using a simple genetic algorithm

This study proposes a method of real-time posture optimization of humanoid robots using a genetic algorithm and neural network. Here, the motion of a humanoid robot pushing an object is considered. When the robot starts pushing the object, the palms of its hands and the soles of its feet are assumed to be fixed on the object and on the ground, respectively, and they sense the reaction force from those surfaces. The reaction force results in changes of torques in the joints. This study determines an optimized posture using a genetic algorithm such that either the torques are evenly distributed over all joints or the torque of the weakest joint is rapidly reduced. Several different optimized postures are then generated by varying the reaction forces at the palms and the soles. The data is used as training patterns for a multilayer perceptron neural network with a back-propagation learning algorithm. Using the trained neural network, the humanoid robot can find the optimal posture for different reaction forces in real time. Several simulations were conducted to confirm the effectiveness of the proposed method. The simulation results showed that the proposed method can be used for real-time posture optimization of humanoid robots.     

基于多类别满意优化控制的磁悬浮球系统

在磁悬浮球控制系统中,针对现有在不同信号输入情况下控制参数优化满意度量化程度低,以及控制系统性能评价指标缺乏针对性的问题,提出多类别满意优化控制算法。即采用基于遗传算法的PID控制器,得到全局中的最满意解,然后选择Kp、Ki、Kd为系统的待优化变量,分别在阶跃、正弦、方波、锯齿以及随机信号的输入下,通过预设的性能评价指标和满意度函数进行分类评价,最终得出满意的优化结果。仿真与实验结果表明,通过多类别满意优化控制算法,不同类别的输入信号都可以在全局中获得控制参数Kp、Ki、Kd的最满意解,最终实现稳定悬浮及动态跟踪的同时给出系统性能评价的量化指标。     

A modular cable-driven humanoid arm with anti-parallelogram mechanisms and Bowden cables

This paper proposes a novel modular cable-driven humanoid arm with anti-parallelogram mechanisms (APMs) and Bowden cables. The lightweight arm realizes the advantage of joint independence and the rational layout of the driving units on the base. First, this paper analyzes the kinematic performance of the APM and uses the rolling motion between two ellipses to approximate a pure-circular-rolling motion. Then, a novel type of one-degree-of-freedom (1-DOF) elbow joint is proposed based on this principle, which is also applied to design the 3-DOF wrist and shoulder joints. Next, Bowden cables are used to connect the joints and their driving units to obtain a modular cable-driven arm with excellent joint independence. After that, both the forward and inverse kinematics of the entire arm are analyzed. Last, a humanoid arm prototype was developed, and the assembly velocity, joint motion performance, joint stiffness, load carrying, typical humanoid arm movements, and repeatability were tested to verify the arm performance.     

形状记忆合金驱动的仿人腕关节机构理论与运动控制

为提高仿人腕关节的运动性能,设计了一款形状记忆合金驱动的仿人腕关节样机,采用拮抗驱动方式实现对仿人腕关节的位置控制,并在相关理论和仿真分析的基础上,通过实验证明该系统模型具有一定的精度。通过样机运动控制实验,研究了拮抗驱动方式对系统运动性能的影响。结果表明,在拮抗驱动方式的不同频率信号跟踪实验中,对于单向正弦信号,跟踪误差最小为[-0.5°, 1°],最大为[-1.5°, 1.5°];对于双向正弦信号,腕关节能实现双向连续偏转,且当信号频率为1/15 Hz时,位置误差均小于[-1.5°, 1.5°]。相比于采用单根形状记忆合金丝驱动,拮抗驱动方式有利于提升位置控制精度和双向偏转能力。     

Effect of driving functions with different jerk and accelerating time length on dynamic performance for mechanical systems: analysis and optimization

We studied the methods that are used for analyzing and optimizing the dynamic performance of mechanical systems in the startup stage under different driving functions. Especially, we considered the effects of joint clearance and component flexibility. The dynamic model of mechanical system is the basis of our methods. A higher order polynomial was used for describing the driving form of mechanical system, and the initial jerk and accelerating time length were employed to determine the specific driving function. Besides, we presented three performance evaluation parameters for mechanical system and introduced surrogate models to improve performance evaluation efficiency. Then, the optimal driving function was determined by using the surrogate models and a multi-objective optimization algorithm. Finally, a planar slider-crank mechanism with a clearance joint was taken as an example to demonstrate our methods.

     

Function-oriented optimization design method for underactuated tendon-driven humanoid prosthetic hand

The loss of hand functions in upper limb amputees severely restricts their mobility in daily life. Wearing a humanoid prosthetic hand would be an effective way of restoring lost hand functions. In a prosthetic hand design, replicating the natural and dexterous grasping functions with a few actuators remains a big challenge. In this study, a function-oriented optimization design (FOD) method is proposed for the design of a tendon-driven humanoid prosthetic hand. An optimization function of different functional conditions of full-phalanx contact, total contact force, and force isotropy was constructed based on the kinetostatic model of a prosthetic finger for the evaluation of grasping performance. Using a genetic algorithm, the optimal geometric parameters of the prosthetic finger could be determined for specific functional requirements. Optimal results reveal that the structure of the prosthetic finger is significantly different when designed for different functional requirements and grasping target sizes. A prosthetic finger was fabricated and tested with grasping experiments. The mean absolute percentage error between the theoretical value and the experimental result is less than 10%, demonstrating that the kinetostatic model of the prosthetic finger is effective and makes the FOD method possible. This study suggests that the FOD method enables the systematic evaluation of grasping performance for prosthetic hands in the design stage, which could improve the design efficiency and help prosthetic hands meet the design requirements.     

Study of the fractional order proportional integral controller for the permanent magnet synchronous motor based on the differential evolution algorithm

A tuning method of the fractional order proportional integral speed controller for a permanent magnet synchronous motor is proposed in this paper. Taking the combination of the integral of time and absolute error and the phase margin as the optimization index, the robustness specification as the constraint condition, the differential evolution algorithm is applied to search the optimal controller parameters. The dynamic response performance and robustness of the obtained optimal controller are verified by motor speed-tracking experiments on the motor speed control platform. Experimental results show that the proposed tuning method can enable the obtained control system to achieve both the optimal dynamic response performance and the robustness to gain variations.     

一种新型拟人机械腿的运动传递性能分析

针对当前多数拟人机械腿采用串联机构出现的运动惯性大、承载能力小,不能很好地满足拟人机器人运动特性需要的问题,提出了一种新型拟人机械腿并联机构,介绍了其结构布局特点,该机构具有3个自由度,外观匀称,与人腿相仿度高,在结构上实现了人体踝关节和膝关节功能.建立了该并联机械腿的位置反解方程,基于该方程用矢量法推导出了线速度和角速度雅可比矩阵,分别定义两矩阵的条件数和可操作度作为其运动学性能指标,分析了各性能指标在工作空间内的分布规律.研究结果表明:拟人机械腿速度传递性能良好且稳定,脚掌的运动传递性能几乎不受屈膝角度的影响.仿真结果符合人体腿部的运动要求,为当前拟人机械腿的研究提供了一种新的构型.