This paper presents a practical adaptive time-delay control scheme (ATDC) and then applies it to robot manipulators. The proposed ATDC uses a time-delayed signal to cancel complex nonlinear dynamics and disturbances, and hence to provide a simple structure. The adaptive law of the proposed ATDC is developed to adjust the control gains appropriately while in operation and then to achieve good tracking performance with tolerant fluctuation and fast convergence speed. Such adaptive control gains are shown to have an effect of adding an adaptive auxiliary control to a conventional time-delay control (TDC). Moreover, the proposed ATDC includes a typical sliding-mode control as a second auxiliary control to suppress the time-delay estimation (TDE) errors caused by the time-delayed signal so that robust performance can be achieved. It is shown by the bounded input bounded output stability theory and a Lyapunov approach that the TDE errors are bounded and the tracking errors are uniformly ultimately bounded, respectively. The practical effectiveness of the proposed ATDC, or the synergistic effect of two auxiliary controls, is illustrated in simulations and experiments with robot manipulators.
This paper presents a widely-adaptive time-delayed control (WATDC) without imposing any restrictions on allowable control gains to make the most of the advantages of time-delayed control (TDC), then applies the WATDC to robot manipulators. Since the proposed WATDC can provide arbitrarily high gains by using the time-varying bounds of time-delayed estimation errors, it has a room for improving the tracking performance and robustness of the existing stabilizing TDCs that have narrow range of control gains depending on system parameters comprising the inertia of a robot manipulator. Furthermore, the proposed WATDC employs a fast adaptive law to generate wide-ranging and well-timed control gains that depend the magnitude of tracking errors. In other words, timely adaptation over a wide range of control gains makes the WATDC provide suitable control effort for desirable output responses. Additionally,to suppress unfavorable inherent time-delayed estimation errors arising from one sample delayed measurements, the proposed WATDC works together with a sliding-mode control that has switching gains based on the time-varying bounds of time-delayed estimation errors. In simulations and experiments with robot manipulators, the proposed WATDC has better tracking performance and robustness than the existing TDCs due to a wide range of adaptive control gains and the time-delayed estimation error dependent switching gains.
An electrochemical lithium-ion battery model is well known to be suited for effectively describing the microstructure evolution in charging and discharging processes of a lithium-ion battery with practically realizable complexity. This paper presents a neural network-based parameter estimation scheme to identify the parameters of an electrochemical lithium-ion battery model in a near-optimal and real-time manner in order to consistently observe the electrochemical states of batteries. The network is first trained to learn the dynamics of the electrochemical lithium-ion battery model, and then, it is applied to estimate the parameters with available finite-time measurements of voltage, current, temperature, and state of charge. In order to efficiently learn the dynamic characteristics of a lithium-ion battery, a well-known recurrent neural network, called a long short-term memory model, is employed with other techniques such as batch normalization, dropout, and stochastic gradient descent with warm restarts for learning speed enhancement and regularization. Using synthetic and experimental data, we show that the proposed estimation scheme works well, finding parameters and recovering the voltage profiles within the root-mean-square error of 0.43% and 26 mV, respectively, even with measurements obtained within a sufficiently short interval of time.