研究発表

This paper presents the estimation of both position and velocity of Stewart Manipulator by means of limbs potentiometer measurements and MEMS inertial sensors. The estimation used the Extended Kalman Filter (EKF) and H∞ optimal filtering technique based on the combination of these sensors. The two types of filters are used as nonlinear state estimators to the Stewart platform which is modeled as a stochastic differential equations due to measurement noise in case of EKF and as a continuous time system model in case of H∞ filtering technique. The results of the both filters are compared with each other on the Stewart platform DELTALAB EX800 using MATLAB SimMechanics toolbox. The simulation results show that Kalman filters are not the best choice for parallel manipulator state estimation as they bear from the hypothesis of statistical noise with zero mean as well as known noise covariance, which may reduce its performance. For these reasons, H∞ filter may be the alternative of Kalman filter.

In this paper, a flocking control algorithm for multiple nonholonomic cars is proposed. First, it is shown that the nonholonomic car is linearized by using a virtual vehicle. Second, we define a metric and topological neighborhood set in order to decide which cars to be next to the other car. Control objectives consist of three rules in boids model as separation, alignment, and cohesion, which is a well-known framework of emerging flock, advocated by Reynolds in 1986. Using the neighborhood set, we show the formation of flocking corresponds to the model. These cars are supposed tobeonastraighthighway,whichenablesustosetanassumptionthateachcarheadstothesamedirectionandreachesthe same velocity eventually. The input for a car consists of the artificial potential method and LaSalle’s invariance principle is utilized to prove the convergence of the entire vehicle system. Finally, the effectiveness of the presented method is examined through simulation.

The purpose of this paper is to propose an optimal traffic signal control in accordance with traffic density to change every moment in traffic network. We focus on calculating optimal signaling split based on traffic volume prediction by model predictive control. Therefore, we aim the achievement of alleviation of congestion in traffic network considering actual traffic situation. Finally the achievement of the purpose is evaluated by controlling of traffic signal in numerical simulations and the relationship between green time and traffic density is examined.

This paper deals with the Simultaneous Localization and Mapping (SLAM) problem for a small Unmanned Aerial Vehicle (UAV) via extended Kalman Filter (EKF) with compensations for unordinary observations. In the SLAM problem, a robot sometimes loses its proper observations then the estimation accuracy deteriorates. In this paper, to remove the effects of unordinary observations, we propose a novel SLAM method considering unordinary observation based on Mahalanobis distance. The proposed method detects the unordinary observations by comparing the observation values with its estimation and determines the weight of these observations. The convergence of the state error covariance matrix is proven. In experimental validation, we show that the UAV state and environment information can be estimated with the proposed method.

Recently, the Aerial Manipulation System becomes very attractive for a wide range of applications due to its unique features. However, control of such system is quite challenging. One of the critical challenge is that this system is verysusceptibletoactuators’faults. Inthispaper,aPassiveFaultTolerantControlSystemisproposedtoaddressthisissue withrobustandoptimalperformance. TherobustnessisachievedusingalinearDisturbanceObserver(DOb)loop. Based on the linearization capability of DOb, a standard Model Predictive Control (MPC) is then used and the resulting control scheme is characterized by both a low computational load and optimal actuators’ efforts with respect to conventional nonlinear robust solutions. This controller is tested to achieve the tracking of a point-to-point task space references under the effect of actuators’ faults, picking/releasing a payload, changing the operating region, and measurement noise. Efficacy of the proposed technique is verified via numerical simulations.

Aerial manipulation systems become very attractive for a wide range of applications due to their unique features. However, control of such system is quite challenging due to its high nonlinearities, couplings, and external disturbances. In this paper, a Disturbance Observer (DOb)-based linearization of a quadrotor manipulation system is utilized. The DOb estimates the disturbances and nonlinearities, then compensates them such that one can treat the control problem based on a simple linear control algorithm. However, the current developed DOb schemes in the literature are based on the precise measurement of the acceleration or the estimation of the velocity. Unlike, these methods, we propose a modified DOb, which is based on both the measured linear accelerations and angular velocities that can be obtained directly from the onboard Inertial Measurement Unit (IMU) and encoders. With this technique, the estimation of nonlinearities and disturbances are carried out without the need of estimation acceleration or velocity, and it is model free. The Effectiveness of the proposed technique is verified via numerical simulations.

In this paper, the load-frequency control problem of microgrid is discussed. The control is applied to maintain a balance between electric-power demand and supply by controlling the outputs of generators in power systems. However, because of the large-scale introduction of energy from renewable-energy sources, it has become more difficult to maintain this balance. To solve this problem, we consider introducing batteries to microgrid. Therefore, we focus on a microgrid that comprises wind-turbine generators, loads, diesel generators, and batteries. First, we design a generalized plant including the diesel generators and batteries. Then, we apply disturbance observer to decide a reference input to diesel generators and batteries. And, we split the reference input into that for diesel generators and that for batteries considering the differences in the response speeds by applying low-pass filter. Furthermore, we apply SoC regulator for the state of charge of batteries reaching maximum capacity or minimum capacity. Finally, we perform numerical simulations to verify that the control method can suppress the frequency deviations and deviations in the residual capacity of the battery.

In this paper, frequency control problem of a power network is discussed. The control is conducted to retain power demandsupplybalancebycontrollingoutputofgeneratorsinpowernetwork. However,itisgettinghardertoretainthebalance because of large amount of renewable energy injection. This paper proposes more economic and ef¿cient frequency control method based on hierarchical model predictive control. At ¿rst, we describe model of the power network. Then, the frequency control based on hierarchical and cooperative model predictive control is proposed. The proposed control method consists of economic dispatch and load frequency control. In numerical simulation, we show the effectiveness of the proposed method comparedwiththeconventionalloadfrequencycontrolmethodintermsofsuppressionoffrequencyvariationandfuelcost.

This paper introduces motion control scheme of a new aerial manipulation system that consists of 2-link manipulator attached to the bottom of a quadrotor. This new system presents a solution for the limitations found in the current quadrotor manipulation systems. It is very attractive for a wide range of applications due to their unique features. However, control of such system is quite challenging because it is naturally unstable, has strong nonlinearities and couplings, has fast dynamics, is very susceptible to parameters variations due to carrying a payload besides the external disturbances like wind, and has actuator limitations. A robust optimal linear control scheme is proposed to address these issues. The proposed control scheme is based on a hybrid linear Model Predictive Control (MPC) and linear Disturbance Observer (DOb) techniques.

—This paper presents a distributed power supplydemand management method based on negawatt trading includingchargeordischargepowerfromenergystoragesystems. In the negawatt trading, power consumers reduce their own power demand and receive monetary reward from the market operator. In order to derive appropriate electricity incentive pricesforconsumersandtouseenergystoragesystemsinpower grids efficiently, we deal with a power adjustment problem using some future power supply information in this paper. We then show a novel market algorithm to solve this problem in a distributed manner through market trading among the market operator and consumers. Finally, we verify the effectiveness of our proposed method with numerical simulation results.

In this paper, we discuss the load-frequency control problem of microgrids. First, we use dispersion-type power sources such as batteries, and secondly, we apply new control theories. Therefore, we focus on a microgrid that comprises a wind-turbine generator, a diesel generator, and a battery, and we propose a technique based on static H∞ control. First, we design a generalized plant considering the differences in the response speeds of diesel generators and batteries. Then, we apply static H∞ control to the generalized plant. Moreover, we combine a linear matrix inequality (LMI) condition with static H∞ control to avoid control gains that will result in control inputs that are too large. We perform numerical simulations to verify that the controller can suppress the frequency deviations and deviations in the residual capacity of the battery. Furthermore, we show that the approach can be employed to design controllers with lower dimensions, and to avoid undesirable control gains. Finally, we perform experiments using a generator and a variable resistance to confirm the effectiveness of the controller.

This paper deals with a regional demand response method based on dynamic electricity pricing in a multi-period energy market. In the proposed method, first, the power demand and supply in each area are determined considering power flow using the retail and wholesale electricity prices in a day-ahead market. Next, to adjust for power deviations caused by errors in power generation, our proposed method solves the problem of optimal allocation between power demand reduction by consumers and increased power supply from balancing generators in a real-time market. This paper also shows a distributed algorithm for obtaining the optimal values of each market player in both the day-ahead and real-time markets, and finally, numerical simulation results demonstrate the effectiveness of the proposed method.

The paper deals with a leader-follower flocking control for a multi-agent system which consists of multiple agents and a leader. The proposed method enables all agents to be disposed at an uniform distance, reach agreement on a same velocity with their neighborhood, and to track a moving leader independently. Leader is seen as one of the agents, in addition, all agents obtain their goal using their neighborhood information only. Furthermore, we suggest the bounded potential function to guarantee the collision avoidance and connectivity maintenance between the agents. As a result, all of the swarm can move with preserving flocking configuration. Finally, numerical simulations show the effectiveness of the proposed method.

This paper deals with the position estimation problem by using the Kalman Filter with compensations for unexpected observations. In the position estimation problem, robot observations sometimes yield unexpected values, resulting in the deterioration of the estimation accuracy. For example, visual observation with an unmanned aerial vehicle often yields unexpected results because of blurred images. In this paper, we propose a method to assigns weights to the observations in order to remove the effects of unexpected observations. In the proposed method, unexpected observations are detected by comparing the observation values with its estimates; the weights of these observations are then determined. On the basis of simulation and experimental results, we demonstrate that a robot’s position can be estimated by the proposed method.

In this paper, the position estimation problem is solved by using unscented Kalman filter with observation uncertainty's compensations. These observation uncertainties causing the estimation become inaccurate in position estimate problem. Visual observation is difficult for an unmanned aerial vehicle equipped with only a monocular-vision camera and often results in observation error because of the detected blurred images. A method to weight the observations is proposed in order to improve the position estimation. Simulation is performed using MATLAB and Simulink to verify the proposed method. The simulation result shows that the proposed method can estimate the position accurately.

This paper deals with a distributed decision procedure for determining the electricity price for a future smart grid. In order to derive the optimal regional prices and improve the convergence speed of the algorithm without private information of market players, we propose a price decision algorithm with an alternating decision making of market players. In this paper, we show that the optimal regional prices derived by using our proposed algorithm maximize the social welfare of power networks which includes profits of power consumers and power suppliers with renewable energy generators. Also, we discuss the convergence property of the proposed price decision algorithm. Finally, the feasibility of our proposed dynamic pricing methodology is shown with numerical simulation results.

In this paper, we propose a cooperative formation control strategy with collision-avoidance capability for a multi-unmanned aerial vehicle (UAV) system by using decentralized model predictive control (MPC) and consensus-based control. Each of the UAVs plans only its own action to track a trajectory. The trajectory is planned for desired formation flying using a consensus-based control algorithm. Collision-avoidance is regarded as coupled constraints. We show how the optimization problems with collision-avoidance can be solved by each UAV in parallel with the other UAVs so that decisions taken independently by each UAV can ensure consistency with the coupled constraints of collision-avoidance. In addition, we show that our proposed approach hardly affects the computation time to solve the optimization problems with collision-avoidance compared to the time for solving the problems without collision-avoidance. Finally, the proposed approach is validated by simulations.