Byoung-Mun Min

Autopilot design of tilt-rotor UAV using particle swarm optimization method

This paper describes an autopilot design of tilt-rotor UAV, which is being developed by KARI as a Smart UAV Development Program in Korea, using particle swarm optimization (PSO) method. The tilt-rotor UAV considered in this paper holds five control modes in the stability and control augmentation system (SCAS) depending on flight mode. Flight control systems designed via the classical approach have been performed in such a way that yields linear models about several trim flight conditions, designing linear controllers for each condition, and integrating these design points with a gain scheduling scheme. However, it is very tedious and time-consuming to design an autopilot of a tilt-rotor UAV which represents various dynamic characteristics, nonlinearity, and uncertainty via classical control technique, because there are many design points and operating conditions throughout the flight envelope. To solve this problem, an automatic tool for control system design using PSO method is developed and applied to autopilot design of tilt-rotor UAV. The desired output of control system is chosen to satisfy the control system requirement. Gain margin and phase margin of control system are additionally considered as a penalty term in the objective function. The designed control system guarantees the satisfaction of the control system requirement ensuring a sufficient stability margin of the control system. Also, the gain scheduling scenario and SCAS switching logic of each control mode are successfully designed. Fully nonlinear 6-DOF simulation for an automatic landing scenario is performed to verify the performance of autopilot system of tilt-rotor UAV. The results from the nonlinear simulation show good control performance of the tilt-rotor UAV.

Guidance Law for Vision-Based Automatic Landing of UAV

In this paper, a guidance law for vision-based automatic landing of unmanned aerial vehicles (UAVs) is proposed. Automatic landing is a challenging but crucial capability for UAVs to achieve a fully autonomous flight. In an autonomous landing maneuver of UAVs, the decision of where to landing and the generation of guidance command to achieve a successful landing are very significant problem. This paper is focused on the design of guidance law applicable to automatic landing problem of fixed-wing UAV and rotary-wing UAV, simultaneously. The proposed guidance law generates acceleration command as a control input which derived from a specified time-to-go (t go ) polynomial function. The coefficient of t go -polynomial function are determined to satisfy some terminal constraints. Nonlinear simulation results using a fixed-wing and rotary-wing UAV models are presented.

Impact angle control guidance law using lyapunov function and PSO method

New guidance law, based on Lyapunov stability theory and applied parameter optimization method, is proposed, which can minimize miss distance and adjust impact angle to predefined value. Using LOS angle and predefine impact angle, we define the state variables and propose a lyapunov function candidate for driving structure of the guidance law and conditions of gain parameters. The PSO algorithm is used for selecting the guidance gains, which satisfies the necessary condition of stability of the guidance loop. This new guidance law has simple structure, shows robustness about different engagement conditions, and provides wider capture area than conventional algorithm considering impact angle only.

Application of Control Allocation Methods to SAT-II UAV

This paper focuses on applying various control allocation schemes to SAT-II UAV system. Control allocation is a useful method for controlling an aircraft with redundant control surfaces and deals with distributing control commands among individual actuators when some control surfaces are failed. In order to implement control allocation scheme in the aircraft control system, control system has to be designed through two step procedures. The first step is to design a baseline control system specifying total control command to be produced, and the second step is to design a control allocator that maps the total control command on individual actuators. The baseline control system of the SAT-II UAV is constituted of speed, altitude, and lateral deviation controllers for tracking a predetermined glide path during automatic landing phase. In the baseline control system, proportionalintegral controllers and lead/lag compensators are effectively combined to assure enough gain and phase margins. Several control allocation methods such as pseudoinverse control allocation method, daisy chain control allocation method, direct control allocation method, and optimization based control allocation methods are considered in this paper. The performance of these methods is evaluated and compared to each other through the nonlinear simulation under the assumption that right aileron failed and are stuck at fixed deflection position during the glide approach phase.

Autonomous flight control system design for a Blended Wing Body

A blended wing body (BWB) UAV has several aerodynamic advantages of lower wetted area to volume ratio and lower interference drag as compared to conventional type UAV. This paper is focused on the design of the autonomous flight control system for a BWB UAV. An onboard control system is developed by using a powerful computer, navigation sensors and communication modem. The autopilot of the BWB UAV is constructed based on the stability analysis using the linearized model at trim condition. We propose a simple control allocation scheme and evaluate its performance through nonlinear simulation. Furthermore, flight test is performed using the designed autopilot and satisfactory performance is obtained in autonomous flight.

Missile autopilot design via output redefinition and gain optimization technique

The autopilot for missile systems should be designed to provide satisfactory stability, performance, and robustness for all flight conditions which may occur in probable engagements. In this paper, dynamic inversion approach based on output redefinition is applied to the missile autopilot design. The redefined output is selected as a linear combination in the ratio of pitch rate and angle-of-attack. Using this redefined output, the proportional and integral gains for the outer-loop autopilot can be systematically determined and, simultaneously, possible to reflect the variation of missile dynamic. However, it is difficult to choice the combination ratio of two state variables, i.e., pitch rate and angle-of-attack, guaranteeing the design criteria of the resulting autopilot. In this paper, a parameter optimization technique is applied to design the autopilot with the dynamic inversion based inner-loop controller using the redefined output. Here, CEALM algorithm is adopted as an optimizer to effectively handle the constraints. Numerical results show that the pitch-channel autopilot designed by the proposed approach presents a satisfactory performance, stability, and robustness against inversion error and parameter uncertainty.

Unmanned autonomous helicopter system design and its flight test

This paper describes an unmanned autonomous helicopter system developed by KAIST UAV team. The developed RUAV (Rotary-Wing Unmanned Aerial Vehicle) system consists of the guidance and control system for autonomous flight, ground control system (GCS), mission payloads, and communication system. The implemented autopilot in RUAV has four control channels for longitudinal and lateral velocities, altitude, and heading angle. A new guidance law was proposed for waypoint navigation and it was slightly modified and applied to various missions. The GCS is composed of three elements such as ground control computer, communication modem, and DGPS base station. The real-time flight data is downloaded to GCS via wireless RF link and stored in GCS. Also, the DGPS correction data, guidance and control commands, operation mode command, and controller gains of autopilot can be uploaded from GCS in real time. The developed RUAV is capable of autonomous take-off and landing and precise hovering. Finally, we demonstrated the performance of our RUAV system through flight test.

Suboptimal guidance laws with terminal jerk constraint

In this paper, a suboptimal guidance law with the terminal constraints on impact angle, acceleration, and time-derivative of acceleration (jerk) is presented. The proposed homing guidance problem is an over-determined problem, i.e., the number of constraints is much larger than the order of system. Since it is difficult to solve the problem by using the conventional optimal control theory, we newly introduce the control input which consists of an additional feed-forward command and the optimal feedback command. The additional feed-forward command is defined as a polynomial function of time-to-go with two degree-of-freedoms. Here, the coefficients of the function in the feed-forward command are determined to satisfy the constraints on terminal acceleration and jerk. The optimal feedback command in consideration of a feed-forward command is determined by Schwartz inequality. The guidance command obtained from the proposed scheme shows a similar form of the biased proportional navigation (BPN) guidance obtained by the optimal control theory. However, the proposed guidance law is less sensitive to the time-to-go estimation error than the BPN law, and it can be easily implemented in the real system. Furthermore, since the coefficients of time-to-go in the feed-forward command remain as design parameters, the proposed guidance law provides more flexibility in application aspect such as trajectory shaping. The characteristics of the proposed guidance law is studied and its performance is also evaluated through numerical simulation.

Analysis on Flight Test Results of Reconfiguration Flight Control System

This paper presents the analysis results obtained by the flight test of reconfiguration flight control system for an aircraft. The reconfiguration flight control system was designed by using control allocation scheme that automatically distributes the demanded control moments determined by control law to each actual control surface. In this paper, some control allocation algorithms for reconfiguration control of general aircraft with redundant control surfaces are summarized and their performance evaluation results through nonlinear simulation and Hardware-In-the-Loop-Simulation (HILS) test are shown. Also, Unmanned Aerial Vehicle (UAV) system adopted as a platform for the flight test of reconfiguration flight controller and the implementation procedure of reconfiguration flight controller into real-time UAV system were introduced. Finally, flight test results were analyzed.

Modeling and Autopilot Design of Blended Wing-Body UAV

This paper describes the modeling and autopilot design procedure of a Blended Wing-Body(BWB) UAV. The BWB UAV is a tailless design that integrates the wing and the fuselage. This configuration shows some aerodynamic advantages of lower wetted area to volume ratio and lower interference drag as compared to conventional type UAV. Also, BWB UAV may be increase payload capacity and flight range. However, despite of these benefits, this type of UAV presents several problems related to flying qualities, stability, and control. In this paper, the detailed modeling procedure of BWB UAV and stability analysis results using the linearized model at trim condition are represented. Finally, we designed the autopilot of BWB UAV based on a simple control allocation scheme and evaluated its performance through nonlinear simulation.