Hagen Daniel A, Marjaninejad Ali, Loeb Gerald E, Valero-Cuevas Francisco J
Department of Biomedical Engineering, University of Southern California, Los Angeles, CA, United States.
Ming Hsieh Department of Electrical and Computer Engineering (Systems), University of Southern California, Los Angeles, CA, United States.
Front Neurorobot. 2021 Oct 11;15:679122. doi: 10.3389/fnbot.2021.679122. eCollection 2021.
Estimates of limb posture are critical for controlling robotic systems. This is generally accomplished with angle sensors at individual joints that simplify control but can complicate mechanical design and robustness. Limb posture should be derivable from each joint's actuator shaft angle but this is problematic for compliant tendon-driven systems where () motors are not placed at the joints and () nonlinear tendon stiffness decouples the relationship between motor and joint angles. Here we propose a novel machine learning algorithm to accurately estimate joint posture during dynamic tasks by limited training of an artificial neural network (ANN) receiving motor angles tendon tensions, analogous to biological muscle and tendon mechanoreceptors. Simulating an inverted pendulum-antagonistically-driven by motors and nonlinearly-elastic tendons-we compare how accurately ANNs estimate joint angles when trained with different sets of non-collocated sensory information generated via random motor-babbling. Cross-validating with new movements, we find that ANNs trained with motor angles tendon tension data predict joint angles more accurately than ANNs trained without tendon tension. Furthermore, these results are robust to changes in network/mechanical hyper-parameters. We conclude that regardless of the tendon properties, actuator behavior, or movement demands, tendon tension information invariably improves joint angle estimates from non-collocated sensory signals.
肢体姿态估计对于控制机器人系统至关重要。这通常通过各个关节处的角度传感器来实现，这种方式简化了控制，但可能使机械设计和鲁棒性变得复杂。肢体姿态应该可以从每个关节的致动器轴角度推导出来，但对于柔顺的腱驱动系统来说这存在问题，在这种系统中，（1）电机不放置在关节处，（2）非线性腱刚度使电机角度和关节角度之间的关系解耦。在此，我们提出一种新颖的机器学习算法，通过对接收电机角度和腱张力的人工神经网络（ANN）进行有限训练，来准确估计动态任务期间的关节姿态，这类似于生物肌肉和腱机械感受器。模拟一个由电机和非线性弹性腱反向驱动的倒立摆，我们比较了通过随机电机乱语生成的不同组非并置感官信息训练时，人工神经网络估计关节角度的准确程度。通过新的动作进行交叉验证，我们发现用电机角度和腱张力数据训练的人工神经网络比没有腱张力训练的人工神经网络更准确地预测关节角度。此外，这些结果对于网络/机械超参数的变化具有鲁棒性。我们得出结论，无论腱的特性、致动器行为或运动需求如何，腱张力信息总是能改善从非并置感官信号得出的关节角度估计。
