Search

This paper gives an overview of classical gainscheduling design applicable to three-loop attitude autopilot. Three-loop autopilot is a classic linear flight control scheme that is still being used nowadays. With the help of classical gainscheduling method it is expected that the implementation can help improve the response of the system. The scopes of this paper include design procedures and analysis from the model linear, design of linear autopilot, until the design of the gainscheduling autopilot. The addition of anti-windup and the effects of discretization and delay are also included in the discussion. The results showed that the designed gain scheduled autopilot is stable in nonlinear model testing and also in digital implementation and the effect of the delay.

This paper proposes the control design ofa nonlinear polynomial fuzzy system with Hâ performance objective using a sum of squares (SOS) approach. Fuzzy model and controller are represented by a polynomial fuzzy model and controller. The design condition is obtained by using polynomial Lyapunov functions that not only guarantee stability but also satisfy the Hâ performance objective. The design condition is represented in terms of an SOS that can be numerically solved via the SOSTOOLS. A simulation study is presented to show the effectiveness of the SOS-based Hâ control designfor nonlinear polynomial fuzzy systems.

In this paper, modelling and simulation of RKX200 rocket flight dynamics are presented. The dynamics model is constructed by first-principle method based on forces and moments acting the rocket rigid body. In the development of the model, the aerodynamics coefficients are determined by using missile DATCOM software. A six Degree of Freedom (DoF) nonlinear model, as well as linear decoupled longitudinal and lateral-directional models are constructed for representing the flight dynamics of the rocket. The resulting models are numerically modelled and simulated using Simulink™ while its motion visualization is constructed using FlightGear™. Further, the dynamics model and the visualization application then are integrated and tested in Hardware In The Loop Simulation (HILS) environment.

The main role of Rocket Navigation System is to deliver an attitude command to Flight Control System hence the rocket can be directed to a particular path or a specific target. This command is produced from Trajectory Planning algorithm. The Navigation System also performs attitude estimation by using the Extended Kalman Filter. The inputs of Navigation System are obtained from GPS, IMU, Pitot tube, Altitude meter and actuator encoder which have been modeled in SIMULINK. From the test results, the Trajectory Planning algorithm is successful enough to give an attitude command and the rocket can reach the target. Also, the Extended Kalman Filter has successfully estimated pitch angle, u-velocity, and w-velocity. Moreover, the Extended Kalman Filter has successfully filtered out the pitch rate noise from the simulated IMU sensor.

Quaternion-based attitude determination system is considered since it can represent all physical attitudes, but not uniquely. An inappropriate quaternion-based control law would imply to a non-efficient control effort and difficulty in guaranteeing global stability. Many research efforts decide to add complexity in the controller by using jump-discontinuity function and/ or hybrid system. In contrast, this paper proposes a continuous quaternion-based attitude control system with an augmented dynamic that guaranteed to be able to regulate any physical state asymptotically. The effectiveness of the designed system is verified through simulations.

One of fixed-wing aircraft type is the fixed-wing unmanned aircraft or fixed-wing Unmanned Aerial Vehicle (UAV). The UAV flies without a pilot in the aircraft. All aircraft movements are controlled by an embedded computer or a remote control. The whole complex UAV control system can be decomposed into several separated sections to simplify the control design process. The three-dimensional position control was simplified to one-dimensional height control and two-dimensional navigation control. The UAV motion composed of three force components and three moment components that make up the six Degrees of Freedom (6 DoF). In the modeling process, all motions of the aircraft are considered linear and have multiple inputs and outputs (MIMO). The controller used was a PID controller which is tuned using Ziegler-Nichols method. Implementation of the Hardware-in-the-Loop-Simulation (HILS) can be done after the design of control systems for UAV is completed. The design of HILS structure was done with the help of MATLAB software. The controller that has been designed previously was implemented into the Ardupilot mega hardware. The design and implementation of PID control system with Ziegler-Nichols tuning method for the longitudinal and lateral directional dimensions, which include angular rate control, attitude control, altitude control and navigation control was able to stabilize an unstable system or to improve the system response.

This paper presents the design and implementation of Three-loop autopilot that functions as an attitude control system of a guided rocket. Three-loop autopilot is designed to manage the guided rocket flight path on a longitudinal plane trajectory by controlling the rocket attitude angle. Three-loop autopilot consists of three successive closed-loop systems which manipulate short period, phugoid oscillation, and other oscillation mode to have adequate damping. Control system is designed on a certain linear condition. Classical stability margin such as gain margin and phase margin is needed to handle error between the assumed model and real system. The system is also equipped with anti-windup to overcome integral windup when the actuator gets saturated. Testing of control system is done by using hardware in the loop simulation (HILS). Control system test shows that the results are good enough. With 0° initial value of pitch and angle of attack, pitch angle can be maintained at a value of 5.5° with a 0.6 seconds rise time, 0.8 peak time, 12% overshoot and zero steady state error. In addition, roll and yaw angle values can be maintained at a value of 0°. On changed command testing, the control system shows good command tracking for small angles, but the system performance decreases when angle value large enough. This is due to the saturation in the actuators and change on system parameters.

These paper present distributed computations of join probabilities SLAM with consensus particle filtering algorithm. We consider groups of robot observe an unknown environment and build the global maps. In this paper, every local map is a global map. Global maps build by using its own information and the information obtained from the other robots. We use particles likelihood as transfer parameters. The information can be transferred to other robot if reach global agreement on particles weight. The global agreement can be obtained by using the Consensus Particle Filtering algorithm, computation is done locally on each robot. After reaching an agreement, then the global map can be built and information of the global map is owned on each robot.

Generally the adjacency matrix is constructed as the function of distance between two nodes in unit disk graph. Although the distance between both robots meet the requirement to form adjacency matrix, however, actually the reliable communication is determined by the function of distance. In this paper, the adjacency matrix that express connectivity in mobile robotic networks are expressed by means of the quality of received wireless signal, which is indicated by received signal strength indicator (RSSI) value. The metric measured in this paper is second semallest eigenvalue or λ2, that represents the connectedness of communication topology in mobile robotics networks. As the results from the simulation, it is shown that the rssi-based adjacency model still satisfy the requirement of connectedness, that is λ2 > 0.

The Ability to extract the maximum power of proton exchange membrane (PEM) fuel cell is the important issue for optimal design of fuel cell powered system. It is necessary to force the fuel cell to operate in a condition which match up with its maximum power point (MPP). MPP varies nonlinearly with the unpredictable variation in the fuel cell's operation condition. Thus, a maximum power point tracking (MPPT) controller is needed to deliver the maximum possible power to the load when a variation in the operation condition occur. The proposed control system consists of two-loop cascade controller with a DC-DC converter. The outer loop uses extremum seeking control to estimate the real-time MPP, and the inner loop uses that estimated value as a set-point for hysteretic control to force the fuel cell to operate at the estimated MPP by DC-DC boost converter. Extremum seeking control and hysteresis control are implemented on a dSPACE DS1104 controller board. For the experiment setup, small scale energy sources are operated, which is Constructor 50 W 10 A fuel cell. Simulation and experiment shows that this proposed control can achieve satisfactorily the maximum power point of fuel cell.

Developing an autonomous flight control system for a fixed-wing unmanned aerial vehicle (UAV) requires the mathematical representation of the system dynamics which can be obtained by applying analytical and/or empirical methods. This paper presents a method for modeling the flight dynamics of the fixed-wing UAV by analytically developing the model structure. The aerodynamic parameters/coefficients of the UAV, to be inserted into the dynamic model, are computed using DATCOM software, by considering various flight condition determined by the angles of attack, altitudes and velocities. In addition to that, an empirical model is also formed based on flight test data, which has been already processed using Kalman filter, to validate the analytical model. By comparing the simulation result of these two systems representations, it can be seen that the both approaches produce dynamic models that have similar characteristics.

Rotation matrix parameterization using unit quaternion can represent attitude globally. However, it is not unique and may generates the unwinding phenomenon, besides the issue on the globally asymptotically stability guarantee. This paper proposes a piecewise-continuous scheme of quaternion-based control law that employs a saturation element such that would avoid the unwinding phenomenon. This paper also proposes a consensus control scheme based on the saturated quaternion control law that would show the property of the unwinding phenomenon avoidance and the corresponding graph only need to have a directed spanning tree.

This paper addresses on the stability analysis of the attitude control system using quaternion-based feedback control law that accommodates arcus-cotangent as a discontinuous elements in the control law. Three argumentations of the asymptotically stability guarantee of the attitude control system are proposed that stated in three theorems. The effectiveness of the proposed control law is shown through numerical simulation.

This paper addresses the static output feedback control synthesis problem for nonlinear polynomial fuzzy systems using a sum of squares (SOS) approach. The open-loop nonlinear systems is represented in the polynomial fuzzy model. By considering the closed-loop system in the output feedback control scheme and the polynomial Lyapunov functions that contain quadratic Lyapunov functions as special cases, sufficient conditions for a solution to the problems of stability analysis and control design can be derived in the representation form of the SOS. It can be numerically solved via the SOSTOOLS.

Spacecraft formation flying control using consensus algorithm means each spacecraft or satellite in a team updates its state based on the information states of its local neighbours in such a way that the final information state of each agent converges to a common value. This paper discusses spacecraft formation within scope of rotational motions i.e. attitude alignment. Two attitude alignment controllers based on consensus algorithm proposed here.Simulations that run as preliminary proof a concept are presented. Simulations run indicate the proposed control laws might solve the attitude alignment problem i.e. consensus is achieved.

In this paper we propose a method of presenting a special case of Value Function as a solution to POMDP in holonomic mobile robot navigation. By using this new method the Value Function complexity will be reduced and more intuitive. The result of this new Value Function is validated with particle filter simulation in Matlab and also experimented physically using a simple autonomous mobile robot built with Lego Mindstorms NXT with 3 ultrasonic sonar and RWTH Mindstorms NXT Toolbox for Matlab to connect the robot to Matlab. This simulation and experiment also incorporate particle filter localization from previous research. The simulation and experiment show that the Value Function can be utilized very well.

In this paper we propose a method of presenting a special case of Value Function as a solution to POMDP in mobile robot navigation. By using this new method the Value Function complexity will be reduced and more intuitive. We also propose a new reinforcement learning method to solve the Value Function. This reinforcement learning is based on Bellman Equation augmented with A* like heuristic during update iteration. The result of this new Value Function is validated with This particle filter is simulaed in Matlab and also experimented physically using a simple autonomous mobile robot built with Lego Mindstorms NXT with 3 ultrasonic sonar and RWTH Mindstorms NXT Toolbox for Matlab to connect the robot to Matlab. This simulation and experiment also incorporate particle filter localization from previous research. The simulation and experiment show that the Value Function can be utilized very well.

This paper present a particle filter also known as Monte Carlo Localization (MCL) to solve the localization problem presented before. A new resampling mechanism is proposed. This new resampling mechanism enables the particle filter to converge quicker and more robust to kidnaping problem. This particle filter is simulated in MATLAB and also experimented physically using a simple autonomous mobile robot built with Lego Mindstorms NXT with 3 ultrasonic sonar and RWTH Mindstorms NXT Toolbox for MATLAB to connect the robot to MATLAB. The particle filter with the new resampling algorithm can perform very well in the physical experiments.

This paper investigates the problem of H∞ performance analysis for continous–time systems with two additive time -varying delays in the state. Our objective is focused on stability analysis of a continuous system with two timevarying delays with an H∞ disturbance attenuation level γ. By exploiting Lyapunov-Krasovski functional and introducing free weighting matrix variables, LMI stability condition have been derived.

This paper present a particle filter for mobile robot localization also known as Monte Carlo Localization (MCL) to solve the localization problem of autonomous mobile robot. A new resampling mechanism is proposed. This new resampling mechanism enables the particle filter to converge quicker and more robust to kidnaping problem. This particle filter is simulated in MATLAB and also experimented physically using a simple autonomous mobile robot built with Lego Mindstorms NXT with 3 ultrasonic sonar and RWTH Mindstorms NXT Toolbox for MATLAB to connect the robot to MATLAB. The particle filter with the new resampling algorithm can perform very well in thesimulation as well as in physical experiments.