• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3416-3435
• /
• 2023

Task offloading in vehicular networks is hot topic in the development of autonomous driving. In these scenarios, due to the role of vehicles and pedestrians, task characteristics are changing constantly. The classical deep learning algorithm always uses a pre-trained neural network to optimize task offloading, which leads to system performance degradation. Therefore, this paper proposes a neural network fine-tuning task offloading algorithm, combining with location prediction for pedestrians and vehicles by the Payne model of fluid dynamics and the car-following model, respectively. After the locations are predicted, characteristics of tasks can be obtained and the neural network will be fine-tuned. Finally, the proposed algorithm continuously predicts task characteristics and fine-tunes a neural network to maintain high system performance and meet low delay requirements. From the simulation results, compared with other algorithms, the proposed algorithm still guarantees a lower task offloading delay, especially when congestion occurs.

• Journal of Electrical Engineering and Technology
• /
• v.11 no.2
• /
• pp.490-498
• /
• 2016

Proportional Integral Derivative (PID) controller tuning is an area of interest for researchers in many areas of science and engineering. This paper presents a new algorithm for PID controller tuning based on a combination of bacteria foraging and particle swarm optimization. BFO algorithm has recently emerged as a very powerful technique for real parameter optimization. To overcome delay in an optimization, combine the features of BFOA and PSO for tuning the PID controller. This new algorithm is proposed to combine both the algorithms to get better optimization values. The real time prototype model of paper machine is designed and controlled by using PIC microcontroller embedded with the programming in C language.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.295-298
• /
• 1996

In this paper, a self-tuning optimal control algorithm is proposed to retain the optimal performance of an active suspension system, when the vehicle has some time varying parameters and parameter uncertainties. We consider a 2 DOF time-varying quarter car model which has the parameter variation of sprung mass, suspension spring constant and suspension damping constant. Instead of solving algebraic riccati equation on line, we propose a neural network approach as an alternative. The optimal feedback gains obtained from the off line computation, according to parameter variations, are used as the neural network training data. When the active suspension system is on, the parameters are identified by the recursive least square method and the trained neural network controller designer finds the proper optimal feedback gains. The simulation results are represented and discussed.

• 제어로봇시스템학회:학술대회논문집
• /
• 1987.10a
• /
• pp.724-728
• /
• 1987

Recently, two degrees of freedom PID control systems have attracted attention, because of their better process control quality for both set point variable tracking and disturbance resistivity than that of conventional PID control systems. This paper describes a new auto-tuning method for the two degrees of freedom PID control system based on a newly developed model matching method in the frequency domain. This new method has been introduced into a two degrees of freedom PID auto-tuning controller, named AdTune TOSDIC-21D8. The superior features of the controller and the results of field tests Are presented.

• Journal of Ocean Engineering and Technology
• /
• v.3 no.2
• /
• pp.551-551
• /
• 1989

In general, a remotely operated vehicle(ROV) operates at deep sea. The control system of ROV is composed of two local loops; the first loop placed on the surface vessel monitors and manipulates the attitude of the ROV using joystick, and the second part on the ROV automatically controls thrusters and acquires positional data. This paper presents a position control simulation of a ROV using an adaptive controller and discusses the control effects of two different conditions. The design of an adaptive control system is obtained by the application of a self-tuning controller with the minimization of an appropriate cost function. The parameters of the control system are estimated by a recursive least square method(RLS). In the simulation, a Runge-Kutta method is used for the numerical integration and the generated outputs are obtained by adding measurement errors. Additionally, this paper discusses the mathematical modelling of a ROV and make a survey of control systems.

• Journal of Ocean Engineering and Technology
• /
• v.3 no.2
• /
• pp.51-58
• /
• 1989

In general, a remotely operated vehicle(ROV) operates at deep sea. The control system of ROV is composed of two local loops; the first loop placed on the surface vessel monitors and manipulates the attitude of the ROV using joystick, and the second part on the ROV automatically controls thrusters and acquires positional data. This paper presents a position control simulation of a ROV using an adaptive controller and discusses the control effects of two different conditions. The design of an adaptive control system is obtained by the application of a self-tuning controller with the minimization of an appropriate cost function. The parameters of the control system are estimated by a recursive least square method(RLS). In the simulation, a Runge-Kutta method is used for the numerical integration and the generated outputs are obtained by adding measurement errors. Additionally, this paper discusses the mathematical modelling of a ROV and make a survey of control systems.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.6 no.3
• /
• pp.40-50
• /
• 1996

By using the neural networks(NN) as system identifier, the on-line self tuning method for fuzzy controller(FC) is proposed. In theis method, the learning of NN is carried out during control operation of FC and the cinsequent parameters of FC is tuned on-line automatically by means of system output errors backpropagated through NN. The Sugeno fuzzy model with constants as consequent parameters is selected for simplifying computation. In procedures of parameter tuning, the gradient descent method is used and the gradient vectors for adjusting the weight of NN are transferred as controller output errors. To evaluate the performance, the proposed method is applied to the inverted pendulum system.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.178.4-178
• /
• 2001

In this paper a self-tuning gain-scheduled skyhook control for semi-active suspension systems is investigated. The dynamic characteristics of a continuously variable damper including electro-hydraulic pressure control valves is analyzed. A 2-d.o.f. time-varying quarter-car model that permits variations in sprung mass and suspension spring coefficient is considered. The self-tuning skyhook control algorithm proposed in this paper requires only the measurement of body acceleration. The absolute velocity of the sprung mass and the relative velocity of the suspension deflection are estimated by using integral filters. The skyhook gains are gain-scheduled in such a way that the body acceleration and the dynamic tire force are optimized. An ECU prototype ...

• 제어로봇시스템학회:학술대회논문집
• /
• 1989.10a
• /
• pp.786-789
• /
• 1989

A nonlinear self-tuning regulator for a neutralization process of a weak acid and strong base system is proposed. Rearranging the state equation of the process model, we first obtain equations which are linear for a manipulated variable or unknown parameters. Then to these equations we apply the standard procedure used in designing linear self-tuning regulators. Simulation results show that the regulator provides very good performances for various realistic situations and traces variations of the unknown parameters. Since computations are simple and additional measurements except the effluent pH value are only flow rates of influent streams, it can be easily applied to real processes such as a waste water treatment process.

• Journal of Mechanical Science and Technology
• /
• v.19 no.9
• /
• pp.1711-1719
• /
• 2005

In an electro statically actuated nanoelectromechanical system (NEMS) resonator, it is shown that both the resonance frequency and the resonance quality (Q) factor can be manipulated. How much the frequency and quality factor can be tuned by excitation voltage and resistance on a doubly-clamped beam resonator is addressed. A mathematical model for investigating the tuning effects is presented. All results are shown based on the feasible dimension of the nanoresonator and appropriate external driving voltage, yielding up to 20 MHz resonance frequency. Such parameter tuning could prove to be a very convenient scheme to actively control the response of NEMS for a variety of applications.