• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.5
• /
• pp.421-426
• /
• 2015

Need to develop human body's posture supervised robots, gave the push to researchers to think over dexterous design of exoskeleton robots. It requires to develop quantitative techniques to assess human motor function and generate the command to assist in compliance with complex human motion. Upper limb rehabilitation robots, are one of those robots. These robots are used for the rehabilitation of patients having movement disorder due to spinal or brain injuries. One aspect that must be fulfilled by these robots, is to cope with uncertainties due to different patients, without significantly degrading the performance. In this paper, we propose chattering free sliding mode control technique for this purpose. This control technique is not only able to handle matched uncertainties due to different patients but also for unmatched as well. Using this technique, patients feel active assistance as they deviate from the desired trajectory. Proposed methodology is implemented on seven degrees of freedom (DOF) upper limb rehabilitation robot. In this robot, shoulder and elbow joints are powered by electric motors while rest of the joints are kept passive. Due to these active joints, robot is able to move in sagittal plane only while abduction and adduction motion in shoulder joint is kept passive. Exoskeleton performance is evaluated experimentally by a neurologically intact subjects while varying the mass properties. Results show effectiveness of proposed control methodology for the given scenario even having 20 % uncertain parameters in system modeling.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.40 no.3
• /
• pp.303-314
• /
• 2020

The hydrological characteristics of watersheds and hydraulic systems of urban and river floods are highly nonlinear and contain uncertain variables. Therefore, the predicted time series of rainfall-runoff data in flood analysis is not suitable for existing neural networks. To overcome the challenge of prediction, a NARX (Nonlinear Autoregressive Exogenous Model), which is a kind of recurrent dynamic neural network that maximizes the learning ability of a neural network, was applied to forecast a flood in real-time. At the same time, NARX has the characteristics of a time-delay neural network. In this study, a hydrological model was constructed for the Taehwa river basin, and the NARX time-delay parameter was adjusted 10 to 120 minutes. As a result, we found that precise prediction is possible as the time-delay parameter was increased by confirming that the NSE increased from 0.530 to 0.988 and the RMSE decreased from 379.9 ㎥/s to 16.1 ㎥/s. The machine learning technique with NARX will contribute to the accurate prediction of flow rate with an unexpected extreme flood condition.

• Journal of Korea Water Resources Association
• /
• v.46 no.4
• /
• pp.373-387
• /
• 2013

For the efficient operation and management of the water resources system, coordinated operation of weirs and reservoirs is required. A simulation based, and an optimization based approaches are available to deal with the operation and management problems. The simulation based approach does not guarantee an optimal solution, and the optimization based approach is not so flexible to consider, complex, nonlinear problems we will face when trying to allocate water to different uses, various demand sectors in a basin. Hence, it is important to develop a model that would compensate for the weak points in both models. We will compare and contrast intrinsic and extrinsic properties of two modeling approaches, addressing issues related to setting system operation and control rules that would lead us to more efficient use of water in the basin. As a result, we propose to use CoWMOM(Coordinated weirs and multi-reservoir operating model), a "simulation based" optimization model for a simple simulation of the past periods, and for the real-time simulation process considering uncertain inflow.