Control of a space robot without actuators on the main body is an underactuated control problem. Various stabilization methods, such as the time-varying feedback control method, discontinuous feedback control method, center manifold-based method, zero-dynamics method and sliding-mode control method have been proposed. However, past studies have not considered underactuated space robots equipped with a flexible appendage, such as solar panels. If the manipulators are simply controlled to achieve the target state for the robot using the past controllers without taking a flexible appendage into consideration, residual vibration remains even after the link motion has finished. In order to suppress the residual vibration on the flexible appendage, we apply the input-shaping technique to the link motion of an underactuated planar space robot. Numerical and experimental studies are carried out to validate the proposed method for a planar dual-link space robot with a flexible appendage. The results show that the proposed method is capable of not only controlling the link angles and the main body attitude to the goal angles, but also suppressing the residual vibration on the flexible appendage.

View full abstract

Flow structures around a NACA0012 wingtip are investigated using a zonal LES/RANS hybrid method in order to understand the noise generation mechanism around the flap-edge. It is known from previous studies that the flow around a blunt wingtip is similar to that of the flap-edge. Grid dependency studies are performed for both time-averaged and unsteady components, and the results are assessed via comparison with the experimental data. It became apparent that the zonal LES/RANS hybrid results are more sensitive to the chordwise grid resolution, compared with these of RANS. Based on the validated data, the noise generation mechanism around the wingtip is discussed. The near-field flow structures relevant to noise generation are obtained successfully.

View full abstract

A receding horizon-based dual control strategy for a planetary landing mission is developed. This strategy introduces the receding horizon framework to solve the nonlinear dynamic path planning problem with the state constraint, which makes up for the defects of the typical polynomial guidance law when it is used in landing on a planet with irregular gravity. Furthermore, the trade-off between efficient control and reliable estimation is considered. The cost incurred by the system uncertainty is incorporated into the performance index. Furthermore a linear feedback control law is provided with the quadratic performance index considering the dual features, which takes advantage of the nonlinear coupling between observability and trajectory to overcome the lack of observability and achieve better estimation performance. By stochastically optimizing the landing trajectory obtained from the receding horizon based convex programming method, the overall performance of the guidance, navigation and control (GNC) system for landing on planets is improved.

View full abstract

The multi-objective genetic algorithm (MOGA) is applied to the multi-disciplinary conceptual design problem for a three-stage launch vehicle (LV) with a hybrid rocket engine (HRE). MOGA is an optimization tool used for multi-objective problems. The parallel coordinate plot (PCP), which is a data mining method, is employed in the post-process in MOGA for design knowledge discovery. A rocket that can deliver observing micro-satellites to the sun-synchronous orbit (SSO) is designed. It consists of an oxidizer tank containing liquid oxidizer, a combustion chamber containing solid fuel, a pressurizing tank and a nozzle. The objective functions considered in this study are to minimize the total mass of the rocket and to maximize the ratio of the payload mass to the total mass. To calculate the thrust and the engine size, the regression rate is estimated based on an empirical model for a paraffin (FT-0070) propellant. Several non-dominated solutions are obtained using MOGA, and design knowledge is discovered for the present hybrid rocket design problem using a PCP analysis. As a result, substantial knowledge on the design of an LV with an HRE is obtained for use in space transportation.

View full abstract

This paper deals with a new lamination theory to calculate the electric current density on carbon fiber reinforced plastic (CFRP) laminates. Unidirectional CFRP has strong orthotropic electric conductance. When electric current is applied to the surface of a CFRP plate, the electric voltage field is not uniform in the thickness direction for thick CFRP. The electric current concentrates near the surface where the electric current is applied to thick CFRP laminates. In this study, a new lamination theory for thick CFRP laminates is proposed. The theory for thick CFRP assumes a non-uniform electric voltage distribution in the thickness direction. For non-thick and non-thin CFRP plates, an approximation method is proposed. To obtain the shape of the non-uniform voltage distribution, the analytical results of thick unidirectional ply from a previous paper is adopted as a contribution function to calculate the effective conductance of the thick CFRP laminate. Cross-sectional two-dimensional (2D) FEM analysis is used to obtain the contribution function for the non-thick CFRP plate. The proposed methods are applied to two cases of the thick CFRP plates, and the results are compared with the three-dimensional (3D) FEM results. Consequently, the new lamination theory is shown to be very effective for the CFRP plates.

View full abstract

During typical supersonic cruising, the temperature of the aircraft skin rises above 300 K due to aerodynamic heating. In this situation, aircraft-skin infrared (IR) suppression, used to minimize the radiation contrast from the background is a crucial survival technology. In the present study, a technique to evaluate the effectiveness of IR suppression of aircraft skin is proposed. For this purpose, a synthetic procedure based on numerical simulations has been developed. In this procedure, the thermal status of aircraft skin is obtained using a computational fluid dynamics (CFD) method for complex aircraft geometries. An IR signature model is proposed using a reverse Monte Carlo (RMC) technique. The detection range and the IR contrast are adopted as the performance indicators for the evaluation of the aircraft IR suppression. The influence of these factors related to the aircraft-skin radiation, such as aircraft-skin emissivity, surface temperature distribution and flight speed, on the IR contrast and the detection range is also studied. As a test case, the effectiveness of various IR suppression schemes was analyzed for a typical air combat situation. Then, the method is applied to clarify the contribution of each aircraft component to the IR suppression of the overall IR radiation. The results show that aircraft-skin temperature control and emissivity control are effective means to reduce the IR radiation and to achieve lower detection. The results can be used as a practical guide for designing future stealth aircraft.

View full abstract

Two spacecraft or more are assumed to be in a state of loose formation flying around a collinear Lagrangian point in the Sun-Earth circular restricted three-body problem (CR3BP) system. The orbit reference of choice for the leader is a halo orbit, and the followers are assumed to follow nearby and be constrained either geometrically or in size. This type of formation could be useful in the future for constructing space ports, space telescopes, astronomical spacecraft requiring sun shields and, with greater numbers, spacecraft swarm missions. The formation design method is constructed by firstly seeking the local coordinate system from the monodromy matrix through extraction of the independent bases that span the space of the halo orbit. To nullify diverging and converging motion, we confine the relative motion to within the periodic subspaces. We observe two modes of relative motion within these subspaces, long-term and short-term motion. In this study, we approximate the long-term motion by deriving a discrete formulation of independent directions based on the eigenvectors of the monodromy matrix, while for the short-term motion we approximate the fundamental set solutions using Fourier series and additional linear functions. Since the size of the formation discussed is significantly smaller than that of the halo orbit, the formation design method can fundamentally be stated as a process of linearly combining these approximations to achieve the desired formation. Consequently, use of this approach transforms formation design from a differential equation problem into an algebraic one, and furthermore enables the long-term and short-term motion design problems to be handled either jointly or separately. A set of design examples is presented to demonstrate the validity of the design method.

View full abstract