利用射频识别(Radio Frequency Identification,RFID)技术对机器人轨迹进行误差测量时,因RFID相位观测值的非连续属性,导致代价函数难以求解,从而使得轨迹误差测量准确性较低。为此,开展了针对RFID智能引导下自动盘点机器人轨迹误差测...利用射频识别(Radio Frequency Identification,RFID)技术对机器人轨迹进行误差测量时,因RFID相位观测值的非连续属性,导致代价函数难以求解,从而使得轨迹误差测量准确性较低。为此,开展了针对RFID智能引导下自动盘点机器人轨迹误差测量技术的研究。在构建的RFID解缠相位-位置模型中,采用相位解缠策略对非连续的RFID相位观测值进行预处理,在包含相位偏移值和相位周期性模糊值的补偿相位辅助下,输出与空间相位值存在平行关系的连续性解缠相位值,解决了非连续属性导致的代价函数难以求解问题,进而实现对机器人位置的精准定位。在此基础上,分别计算机器人当前位置在x、y方向上的误差,并通过微分操作,输出连续的轨迹误差。在100 m×80 m的实验环境中,所提技术可以精准定位机器人位置,其在x方向轨迹误差测量偏差≤0.5 cm,y方向偏差≤0.1 cm,显著提升了轨迹误差测量的准确性与稳定性。因此,研究为RFID引导下机器人定位与轨迹精度评估提供了一种可靠的技术途径,对仓储自动化、机器人智能巡检等应用具有实际意义。展开更多
Phase measurement profilometry(PMP) uses a digital projector and a camera for 3D shape measurement.However,the nonlinear response of the measurement system causes the captured perfect sinusoidal fringe patterns to bec...Phase measurement profilometry(PMP) uses a digital projector and a camera for 3D shape measurement.However,the nonlinear response of the measurement system causes the captured perfect sinusoidal fringe patterns to become non-sinusoidal waveforms,which results in phase and measurement errors.We perform a theoretical analysis of the phase error resulting from non-sinusoidal fringe patterns.Based on a derived phase-error expression,the empirical mode decomposition(EMD) method is introduced to restrain nonlinear phase error and improve the precision in evaluating the phase distribution.A computer simulation and experimental results prove that the proposed method can eliminate possible phase-error in PMP.展开更多
基金sipported by the National Natural Science Foundation of China (61128012,60978043,61061160503)the Research Grants Council of the Hong Kong Special Administrative Region (9041577)
文摘Phase measurement profilometry(PMP) uses a digital projector and a camera for 3D shape measurement.However,the nonlinear response of the measurement system causes the captured perfect sinusoidal fringe patterns to become non-sinusoidal waveforms,which results in phase and measurement errors.We perform a theoretical analysis of the phase error resulting from non-sinusoidal fringe patterns.Based on a derived phase-error expression,the empirical mode decomposition(EMD) method is introduced to restrain nonlinear phase error and improve the precision in evaluating the phase distribution.A computer simulation and experimental results prove that the proposed method can eliminate possible phase-error in PMP.
