基于H∞的双足机器人关节控制方法研究

本文利用拉格朗日法建立了双足步行机器人的动力学模型,从中提取出干扰和摄动.通过分析和求解H∞状态反馈控制问题,设计了一个鲁棒控制器,该控制器既能使双足步行机器人控制系统对干扰不敏感,又对摄动保持稳定.得到H∞控制器后,给出了双足步行机器人的全局关节控制系统,并且对双足步行机器人进行了仿真实验,实验结果验证了控制方法的有效性.     

仿人预测控制在双足机器人步行控制中的应用

基于零力矩点(ZMP)的预测控制是目前双足机器人步行控制中最先进的方法,但是预测控制需要比较精确的预测模型,在环境扰动导致模型失配时,预测控制的性能下降较快。为了解决这个问题,利用仿人智能控制对环境误差具有较强抑制的特点改进预测控制。探讨了在步行控制中引入仿人智能控制的必要性和仿人智能控制改进预测控制的可行性,并设计了仿人预测控制器。最后通过仿真实验验证了新的控制器对双足机器人步行控制的有效性。     

学习人类控制策略的双足机器人步态控制研究

针对双足机器人面临的复杂环境下动态行走的适应性难题,提出了一种基于学习人类控制策略的双足机器人步态控制方法。利用三维线性倒立摆模型构造双足行走系统的状态方程,建立学习人类控制策略的参数化模型,设计了基于SVM的学习型控制器。该方法保证了躯干始终处于与地面近似垂直,增强了步态控制的鲁棒性,提高了双足机器人在复杂环境下行走的动态稳定性。实验验证了该方法的有效性。     

基于全身协调的仿人机器人步行稳定控制

提出利用机器人质心（CoM）雅克比矩阵,实现全身协调补偿的算法。提出机器人的简化模型;分析基于CoM雅克比矩阵的补偿算法;采用CoM/ZMP（零点矩点）、减振和软着陆控制器实时控制双足步行,实现机器人全身协调的稳定控制;通过仿人机器人AFU09的双足步行实验证明该控制方法的有效性。     

基于降阶位置/力模型的机器人神经网络控制

双足机器人的双脚支撑期是实现其步行运动的重要过程,然而耦合的位置/力控制难以保证其稳定平滑运动.本文提出了一种基于降阶位置/力模型的机器人控制策略,整合了位置控制子空间模型和力控制子空间模型,通过模型降阶减小了控制器设计的复杂度,并采用神经网络自适应控制方法综合多控制目标,实现了双足机器人的平滑稳定控制并有效地抑制了系统外扰和参数不确定性的影响.最后,仿真算法验证了该控制方法和模型的有效性.     

双足机器人上楼梯步态的规划与控制

研究双足机器人上楼路径优化技术,为了让双足机器人实现稳定的上楼梯的行走,应规划双足机器人的上楼梯的步态.为解决对步态进行合理规划与稳定控制问题,提出用几何约束法,用摆线拟合踝关节运动轨迹,设计模糊控制器,对踝关节的滚转角度进行调整,使ZMP位置靠近支撑区域中心,保证了机器人的稳定行走;最后在ADAMS软件中建立了双足机器人的虚拟样机,并通过与Matlab的联合仿真,实现了双足机器人上楼梯的稳定行走仿真.仿真结果验证了上楼梯步态与模糊控制器的有效性,为系统设计提供了保证.     

仿人机器人相似性运动轨迹跟踪控制研究

提出一种基于带观测器的条件状态反馈控制的仿人机器人相似性运动轨迹跟踪控制方法.首先,分析了7连杆双足机器人动力学模型,阐述了其运动能量方程与动力学特征方程; 其次,基于带观测器的状态反馈控制器原理,构建起三维倒立摆平衡控制模型; 最后,由线性二次型调节器确定状态反馈增益矩阵,使机器人轨迹跟踪误差最小化,以复现出较高相似度的双足步行运动效果.实验验证了该方法的有效性.     

双足机器人的鲁棒控制

利用拉格朗日法建立了双足机器人的动力学模型．在双脚支撑地时，运动学方程的约束造成双足机器人自由度的冗余，本文引入拉格朗日因子消除了双足机器人的冗余自由度．采用鲁棒控制法对双足机器人的轨迹跟踪进行控制，仿真实验结果证明, 鲁棒控制法对模型不精确或外部干扰对双足机器人产生的影响有很好的抑制作用，对双足机器人轨迹跟踪控制是有效的．     

双足步行机器人能量成型控制

为了使双足被动行走机器人的行走步态符合仿生规律,且当路面坡度变化后,迅速进入新的稳定步态行走,提出了角度不变能量成型控制策略.研究了欠驱动双足机器人能量匹配条件和能量成型控制器的求解;由于动能相对于旋转变换不具有对称性,通过在能量成型控制中附加一个辅助控制量,实现角度不变控制.仿真结果表明,该算方法可实现仿生控制,既能扩大吸引域,又改善系统的鲁棒性.     

双足机器人步态控制研究方法综述

概括地介绍了双足机器人步态控制领域内的主要研究思路．详细阐述了基于双足动力学特征的3种建模方法,包括倒立摆模型、被动步态模型、质量弹簧模型的特点．另外讨论了两种常用的约束条件(稳定判据与能量约束)和3种智能控制方法(神经元理论、模糊逻辑与遗传算法)在双足机器人步态控制中的研究情况．     

Robust Sliding-mode Control of Nine-link Biped Robot Walking

A nine-link planar biped robot model is considered which, in addition tothe main links (i.e., legs, thighs and trunk), includes a two-segment foot.First, a continuous walking pattern of the biped on a flat terrain issynthesized, and the corresponding desired trajectories of the robot jointsare calculated. Next, the kinematic and dynamic equations that describe itslocomotion during the various walking phases are briefly presented. Finally,a nonlinear robust control approach is followed, motivated by the fact thatthe control which has to guarantee the stability of the biped robot musttake into account its exact nonlinear dynamics. However, an accurate modelof the biped robot is not available in practice, due to the existence ofuncertainties of various kinds such as unmodeled dynamics and parameterinaccuracies. Therefore, under the assumption that the estimation error onthe unknown (probably time-varying) parameters is bounded by a givenfunction, a sliding-mode controller is applied, which provides a successfulway to preserve stability and achieve good performance, despite the presenceof strong modeling imprecisions or uncertainties. The paper includes a setof representative simulation results that demonstrate the very good behaviorof the sliding-mode robust biped controller.     

基于SERCOS总线的两足步行机器人分布式运动控制系统

本文介绍两足步行机器人运动控制系统结构的新发展,并介绍一种适于多轴运动控制 的通信系统——SERCOS总线,它的通信速率高,实时性和确定性好,很适合机器人各关节的 协调运动控制,便于构成分布式的机器人控制系统结构．文中讨论了基于SERCOS总线的两足 步行机器人控制系统结构设计,并分析了其优点．     

Walking Control Algorithm of Biped Humanoid Robot on Uneven and Inclined Floor

This paper describes walking control algorithm for the stable walking of a biped humanoid robot on an uneven and inclined floor. Many walking control techniques have been developed based on the assumption that the walking surface is perfectly flat with no inclination. Accordingly, most biped humanoid robots have performed dynamic walking on well designed flat floors. In reality, however, a typical room floor that appears to be flat has local and global inclinations of about 2°. It is important to note that even slight unevenness of a floor can cause serious instability in biped walking robots. In this paper, the authors propose an online control algorithm that considers local and global inclinations of the floor by which a biped humanoid robot can adapt to the floor conditions. For walking motions, a suitable walking pattern was designed first. Online controllers were then developed and activated in suitable periods during a walking cycle. The walking control algorithm was successfully tested and proved through walking experiments on an uneven and inclined floor using KHR-2 (KAIST Humanoid robot-2), a test robot platform of our biped humanoid robot, HUBO.     

欠驱动两足步行机器人研究现状

针对欠驱动两足步行机器人的研究现状与发展趋势进行了探讨。首先,总结了被动行走和踝关节欠驱动两足机器人的研究现状,介绍了欠驱动两足步行机器人的基本研究方法,包括问题描述、步态规划、运动控制和稳定性判定等,并对欠驱动两足机器人需要进一步研究的问题和发展方向进行深入研究,最终目标是将欠驱动控制策略应用于两足步行机器人的行走过程控制,以提高其运动性能。     

一种小样本支持向量机控制器在两足机器人步态控制的研究

神经网络等传统的机器学习方法是基于样本数目无穷大的经验风险最小化原则,这对非确定环境下有限样本的步态学习控制非常不利．针对两足机器人面临的非确定环境适应性难题,提出了一种基于支持向量机（SVM）的两足机器人步态控制方法,解决了小样本条件下的步态学习控制问题．提出了一种基于混合核的步态回归方法,仿真研究表明了这种方法比全局核和局部核分别单独用于步态学习时有优越性．SVM以踝关节及髋关节的轨迹作为输入,相应的满足ZMP判据的上体轨迹作为输出,利用有限的理想步态样本对机器人上体轨迹与腿部轨迹之间的动态运动关系进行学习,然后将训练好的SVM置入机器人控制系统,从而增强了步态控制的鲁棒性,有利于实现两足机器人在非结构环境下的稳定步行．仿真结果表明了所提方法的优越性．     

Stability analysis via the concept of Lyapunov exponents: a case study in optimal controlled biped standing

Balancing control is important for biped standing. In spite of large efforts, it is very difficult to design balancing control strategies satisfying three requirements simultaneously: maintaining postural stability, improving energy efficiency and satisfying the constraints between the biped feet and the ground. In this article, a proportional-derivative (PD) controller is proposed for a standing biped, which is simplified as a two-link inverted pendulum with one additional rigid foot-link. The genetic algorithm (GA) is used to search for the control gain meeting all three requirements. The stability analysis of such a deterministic biped control system is carried out using the concept of Lyapunov exponents (LEs), based on which, the system stability, where the disturbance comes from the initial states, and the structural stability, where the disturbance comes from the PD gains, are examined quantitively in terms of stability region. This article contributes to the biped balancing control, more significantly, the method shown in the studied case of biped provides a general framework of systematic stability analysis for certain deterministic nonlinear dynamical systems.     

A binary approximating method for graspable region determination of biped climbing robots

For a biped pole-climbing robot (BiPCR) with grippers, it is an essential demand to determine the target grasp configuration for climbing and transiting between poles, with the graspable region as a priori knowledge. The graspable region on the target pole is critically important for climbing path planning and motion control. To efficiently compute the graspable region for a BiPCR, we propose a novel binary approximating method in this paper. This method may also be applied to generate the three-dimensional (3-D) workspace of a manipulator with constant orientation. The grasping problem and the concept of graspable region for a BiPCR are first introduced. The binary approximating method and the corresponding algorithms are then presented to generate the graspable region. Additional constraints on a biped climbing robot with five degrees of freedom (DoFs) are presented as a supplement to the algorithm. A series of comprehensive simulations are conducted with the five-DoF and six-DoF climbing robots to verify the effectiveness of the proposed method. Finally, the dexterity of biped climbing robots with different DoFs is discussed.     

基于deep Q-network双足机器人非平整地面行走稳定性控制方法

针对双足机器人在非平整地面行走时容易失去运动稳定性的问题,提出一种基于一种基于价值的深度强化学习算法DQN（Deep Q-Network）的步态控制方法。首先通过机器人步态规划得到针对平整地面环境的离线步态,然后将双足机器人视为一个智能体,建立机器人环境空间、状态空间、动作空间及奖惩机制,该过程与传统控制方法相比无需复杂的动力学建模过程,最后经过多回合训练使双足机器人学会在不平整地面进行姿态调整,保证行走稳定性。在V-Rep仿真环境中进行了算法验证,双足机器人在非平整地面行走过程中,通过DQN步态调整学习算法,姿态角度波动范围在3°以内,结果表明双足机器人行走稳定性得到明显改善,实现了机器人的姿态调整行为学习,证明了该方法的有效性。