A Novel Walking Speed Control Strategy for Power Augmentation Exoskeletons based on Neural Networks
Abstract
This paper addresses a novel sensing technique to minimize the interaction force of walking speed transitions during the navigation of a coupled human-exoskeleton power augmentation system. The proposed technique is able to classify the intended walking speed based on Dual Reaction Force (DRF) Sensor. The human brain as a complex information processing device is quite difficult to be simulated, especially when considering the processing speed and the speed of sensed signals through the human nervous system throughout the human body. We developed the DRF sensors for preemptive identification of pilot intentions for walking speed transitions to augment the response of the exoskeleton to shadow pilot's movements. The Dynamic Threshold Neural Networks (DTNNs) is used to classify the input signals and make a decision on the transition of system's walking speed according to the pilot's intentions and walking speed limitation. The actions for walking speed transitions are simulated in MATLAB/Simulink, and Variable Admittance Controller (VAC) and applied. To show the efficiency of the proposed walking speed control strategy, comparison is conducted with ordinary VAC algorithm technique.
References
Ka M, Cheng H, Toan T H, Jing Q. “Minimizing Human-Exoskeleton Interaction Force Using Compensation for Dynamic Uncertainty Error with Adaptive RBF Network” , Journal of Intelligent and Robotic Systems, 2015, 82: 413–-433.
H. Kazerooni, A. Chu, R. Steger, ``That Which Does Not Stabilize, Will Only Make Us Stronger'', International Journal of Robotics Research, vol. 26, 2007, pp. 75-89.
A. Zoss , H. Kazerooni, A. Chu, ``On the Mechanical Design of the Berkeley Lower Extremity Exoskeleton (BLEEX)”,International Conference on Intelligent Robots and Systems, Edmenton, Canada, 2005, pp. 3132-3139.
A. I. A. Ahmed, H. Cheng, Z. Liangwei, M. Omer, X. Lin, “On-line Walking Speed Control in Human-Powered Exoskeleton Systems Based on Dual Reaction Force Sensors”, Journal of Intelligent and Robotic Systems, vol. 87 , no.1 , 2017, pp. 59-80.
H. Kazerooni, J. L. Racine, L. Huang, R. Steger, “On the Control of the Berkeley Lower Extremity Exoskeleton (BLEEX)”, International Conference of Robotics and Automation, Barcelona, Spain, 2005, pp. 4353-4360.
J. Ghan, R. Steger, H. Kazerooni, “Control and System Identification for the Berkeley Lower Extremity Exoskeleton”, Advanced Robotics, vol. 20, 2006, pp. 989-1014.
Y. Sankai, ``HAL: Hybrid Assistive Limb based on Cybernics”, Robotics Research, 2011, pp. 25-34.
S. Lee, Y. Sankai, “Virtual Impedance Adjustment in Unconstrained Motion for an Exoskeletal Robot Assisting the Lower Limb", Advanced Robotics, vol. 19, 2005, pp. 73-795.
Okunev V., Nierhoff T., H.S.: Human-preference-based control design : Adaptive robot admittance control for physical human-robot interaction. In: The 21st IEEE International Symposium on Robot and Human Interactive Communication, 2012, pp. 443-448.
Kim J W, Jang H J, Hwang D H, Park C., “A Step, Stride and Heading Determination for the Pedestrian Navigation System”, Journal of Global Positioning Systems, vol. 3, 2004, pp. 273-279.
Righetti L, Buchli J, Ijspeert A J, “Dynamic hebbian learning in adaptive frequency oscillators” Physica D: Nonlinear Phenomena, vol. 216, 2005, pp. 269-281.
Isakov E, Burger H, Krajnik J, Grgoric M, Marincek C., “Influence of speed on gait parameters and on symmetry in transtibial amputees”, Prosthetics and Orthotics International, vol. 20, 1996, pp. 153-158.
Mondal S, Nandy A, Verma C, Shukla S, Saxena N, Chakraborty P, Nandi G C., “Modeling a Central Pattern Generator to Generate the Biped Locomotion of a Bipedal Robot Using Rayleigh Oscillators”, Communications in Computer and Information Science , vol. 168, 2011, pp. 289-300.
Hornik K, Stinchcombe M, White H., “Multilayer feedforward networks are universal approximators”, Neural Networks, 1989, vol.2, pp. 359-366.
Chen T P, Chen H., “Universal approximation to nonlinear operators by neural networks with arbitrary activation functions and its application to dynamical systems”, IEEE Transactions on Neural Networks, 1995, vol. 6, pp. 911–917.
S. Lee, Y. Sankai, “Power Assist Control for Walking Aid with HAL-3 Based on EMG and Impedance Adjustment around Knee Joint”, International Conference on Intelligent Robots and Systems, EPFL, Switzerland, 2002, pp. 1499-1504.
A. I. A. Ahmed, H. Cheng, X. Lin, Z. Elhassan, M. Omer, “On-line Walking Speed Control in Human-Powered Exoskeleton Systems”, IEEE International Conference on Communication, Control, Computing and Electronic Engineering(ICCCCEE2017), Khartoum, Sudan, 2017, pp.1-7.
Tanvi S, Arpit A, Annu B, Mahesh C.,”SVM and ANN: A Comparative Evaluation”, 1st International Conference on Next Generation Computing Technologies (NGCT-2015), Dehradun, Uttrakhand, India, 2015, pp. 960-964.
Roman A P., “Log-Sigmoid Multipliers Method in Constrained Optimization”, Annals of Operations Research, 2001, vol. 101, pp. 427–460.
Zhang C, Woodland P C., “Parameterised Sigmoid and ReLU Hidden Activation Functions for DNN Acoustic Modelling”, IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Shanghai, China, 2016, pp. 5300-5304.
Abusabah I. A. Ahmed, H. Cheng, A. R. Abdalla and G. Hongliang, "Modified admittance control for maneuverable human-powered augmentation lower exoskeleton systems”, 2017 2nd International Conference on Advanced Robotics and Mechatronics (ICARM), Hefei, 2017, pp. 7-12.
