Institute of Mechanical Engineering, National Chiao-Tung University, Taipei, Taiwan, ROC.
Gait Posture. 2011 Apr;33(4):701-5. doi: 10.1016/j.gaitpost.2011.03.005. Epub 2011 Apr 1.
In human motion analysis, in situ calibration of the force plate is necessary to improve the accuracy of the measured ground reaction force (GRF) and center of pressure (COP). Few existing devices are capable of both static and dynamic calibration of the usually non-linear GRF and COP errors, while are also easy to move and/or set up without damaging the building. The current study developed a small device (160 cm × 88 cm × 43 cm) with a mass of 50 kg, equipped with auxiliary wheels and fixing suction pads for rapid deployment and easy set-up. A PC-based controller enabled quick movement and accurate positioning of the applied force to the calibration point. Static calibration at 100 validation points and dynamic calibration of a force plate were performed using the device. After correction by an artificial neural network (ANN) trained with the static data from another 121 points, the mean errors for the GRF were all reduced from a maximum of 0.64% to less than 0.01%, while those for the COP were all reduced from a maximum of about 1.37 mm to less than 0.04 mm. For dynamic calibration, the mean errors for the GRF were reduced from a maximum of 0.46% to less than 0.28%, while those for the COP were reduced from a maximum of 0.95 mm to less than 0.11 mm. The results suggest that the calibration device with the ANN method will be useful for obtaining more accurate GRF and COP measurements in human motion analysis.
在人体运动分析中，为了提高测量地面反作用力（GRF）和压力中心（COP）的准确性，有必要对力板进行原位校准。现有的一些设备能够对通常是非线性的 GRF 和 COP 误差进行静态和动态校准，同时易于移动和/或设置，而不会损坏建筑物。本研究开发了一种小型设备（160cm×88cm×43cm），重 50kg，配备辅助轮和固定吸盘，可快速部署和轻松设置。基于 PC 的控制器可实现施加力到校准点的快速移动和精确定位。使用该设备对 100 个验证点进行静态校准，并对力板进行动态校准。通过用另一个 121 个点的静态数据训练人工神经网络（ANN）进行校正后，GRF 的平均误差从最大 0.64%降低到小于 0.01%，而 COP 的平均误差从最大约 1.37mm 降低到小于 0.04mm。对于动态校准，GRF 的平均误差从最大 0.46%降低到小于 0.28%，而 COP 的平均误差从最大 0.95mm 降低到小于 0.11mm。结果表明，具有 ANN 方法的校准设备将有助于在人体运动分析中获得更准确的 GRF 和 COP 测量值。
