【作者】 刘鹏；

【导师】 苏春翌； 陈春玉；

【作者基本信息】 广东工业大学 ， 控制工程（专业学位）， 2025， 硕士

【摘要】 在机器人技术不断发展的当下,机器人在复杂任务中的应用日益增多,变阻抗控制技术作为提升其适应性和交互性能的关键,备受关注。然而,变阻抗控制面临的一个重大挑战是,它有可能破坏机器人的无源性,而无源性是确保机器人稳定性与安全性的核心保障。针对这一问题,本文进行了基于能量罐的机器人变阻抗控制研究。通过理论分析与实验验证相结合的方式,对所提出的控制策略进行了全面深入的探讨。本文采用7-DOF Franka Emika Panda冗余机械臂作为实验平台,并在第三、四、五章中基于该机械臂完成了实机测试,旨在验证所提出控制方法的有效性和可靠性。本文的主要研究工作如下:(1)针对经典能量罐无法严格限制能量罐上界的问题,本文设计了一种基于障碍函数的能量罐。通过引入障碍函数,能够有效地预先确定并维持能量罐的能量上限,从而克服了传统能量罐在能量约束方面的不足。在此基础上,本文进一步引入复合误差的概念,构建了系统能量函数。借助该能量函数,以此证明能量罐的有界性以及系统的无源性。在此基础上,本文提出了一种保持无源特性的可变阻抗控制器。该控制器能够确保系统的无源性,从而有效防止机器人出现不稳定和不安全的行为,为机器人的稳定运行和安全操作提供了有力保障。(2)传统的能量罐在与机器人交换能量时通常采用硬切换机制,这会导致抖动问题,显著影响系统的稳定性与操作精度。为克服这一局限,本文引入了一种新型能量罐框架,采用平滑切换函数,并提出了一种新的可变阻抗控制策略。该框架的关键优势在于能够在保持系统无源性的同时实现能量的平滑过渡,有效抑制因硬切换导致的抖动。此外,本文还引入了-修正技术,确保能量罐保持有界,而无需局限于固定的范围,从而为系统的稳定运行提供更大的灵活性和适应性。(3)针对由于机械臂运动过程中接触点丢失而产生的接触性损失问题,使得系统短时间内从能量罐中提取大量能量,从而出现危险行为的问题,本文提出了一种基于安全阀的能量罐自适应力位混合控制方法。该方法通过安全阀调节从能量罐到系统的功率流,并设计了自适应函数,使机器人能够根据系统状态动态调整自身的刚度和力控制器的大小。这种设计不仅提高了机器人系统在复杂任务中的稳定性和安全性,还在人机交互过程中表现出更加出色的柔顺性。更多还原

【Abstract】 In the rapidly advancing field of robotics,robots are increasingly being applied to complex tasks.Variable impedance control technology,which enhances their adaptability and interac-tive performance,has garnered significant attention.However,a major challenge in variable impedance control is its potential to compromise the passivity of robots,which is essential for ensuring their stability and safety.To address this issue,this thesis proposes a variable impedance control method for robots based on the energy tank.By combining theoretical analysis with experimental validation,the proposed control strategy is comprehensively ex-plored.The 7-DOF Franka Emika Panda robotic manipulator was selected as the experimental platform,and the experiments in Chapters 3,4,and 5 were all conducted on this robotic ma-nipulator to verify the effectiveness and reliability of the proposed control method.The main contributions of this thesis are as follows:(1)To address the limitation of traditional energy tanks that cannot strictly constrain the upper energy bound,this thesis proposes an energy tank design based on a barrier function.By introducing the barrier function,the upper energy bound of the tank can be effectively prede-termined and maintained,overcoming the constraints of conventional energy tanks.Building on this foundation,the thesis introduces the concept of composite error to construct a system energy function.This function is used to demonstrate the boundedness of the energy tank and the passivity of the system.Furthermore,a variable impedance controller that preserves pas-sivity is proposed.This controller ensures system passivity,effectively preventing instability and unsafe behavior in robots,thereby providing robust guarantees for stable robot operation and safe human-robot interaction.(2)Traditional energy tanks typically employ a hardswitching mechanism when exchang-ing energy with robots,leading to chattering issue,which signiffcantly compromises stability and manipulation accuracy.To overcome this limitation,the thesis introduce a novel energy tank framework using smooth switching functions and propose a new variable impedance con-trol strategy.The key advantage of this framework is its ability to achieve smooth energy transitions while maintaining system passivity,effectively suppress chattering caused by hard switching.Additionally,the thesis introduce the-modiffcation technique,which ensures that the energy tank remains bounded without being constrained to a ffxed range,thereby providing greater ffexibility and adaptability for stable system operation.(3)To address the issue of contact loss caused by the loss of contact points during robotic manipulators motion,which leads to rapid energy extraction from the energy tank and poten-tially hazardous behavior,this thesis proposes an adaptive force/position hybrid control method for the energy tank based on a safety valve.The method regulates the power flow from the en-ergy tank to the system through the safety valve and designs an adaptive function,enabling the robotic arm to automatically adjust its stiffness and the magnitude of the force controller according to the current system state.This design not only enhances the stability and safety of the robotic system in complex tasks but also demonstrates superior compliance during human-robot interaction.更多还原