【摘要】 类人机器人灵巧手抓取过程中的惯性力会影响到抓取的鲁棒性,为了消除惯性力的影响,需要检测出加在手上的空间加速度反馈给控制系统,以确保抓取的平稳性和快速性。采用并联机构作为弹性元件研究了一种六维加速度传感器,建立了静态数学模型,推导了加速度和灵敏度雅可比矩阵,并对传感器的静态特性进行了研究。根据定义的静态性能指标采用空间模型理论的图谱寻优法,优化设计了传感器的理论模型参数。为了验证设计参数的可靠性,给出了1种类人机器人腕用六维加速度传感器的算例,仿真实验获得的各弹性连接杆上应变的实验值与理论值的最大误差为1.617%。最后标定实验结果证明研究的理论和方法是有效和可靠的。更多还原

【Abstract】 The acceleration of humanoid robot wrist disturbs the grasp robustness of the multi-fingered robot hand.In order to acquire the acceleration,effectively counteract the disturbance caused by the acceleration and ensure robot hand grasping stably and accurately,this paper presents a six-axis accelerometer that uses a parallel mechanism as the sensing element for humanoid robot wrist,establishes its mathematics model,deduces the acceleration Jacobian matrix and sensitivity Jacobian matrix and analyses its static characteristics.Based on the defined static characteristic indices,the theoretical model parameters are optimized using the physical model of solution space.The computing example of a six-axis accelerometer for humanoid robot wrist is given;the acquired maximum error between the theoretical strains and experimental strains of each leg is 1.617%,which verifies that the design parameters are valid.Finally,the results of calibration experiment also testify that the proposed theory and method are valid and reliable.更多还原