An unique automatic truss beam builder was developed in this study, whose functions can be verified easily at the gravity condition on the ground. This beam builder has vertically symmetric mechanisms, and extends two octahedral truss beams simultaneously by symmetric motions. The energy loss against the gravity and the gain from the gravity are compensated via inner forces of the mechanical elements. This condition is similar to the weightlessness in terms of the energy consumption. The advantage of this compensation method is that this beam builder compensates the energy cansumption due to the gravity by simple mechanisms by itself, and requires no other complex facilities, such as a suspension system by tension controlled cables or a neutral buoyancy pool. When this method is applied to developing and verifying space robots, the energy consumption and the working speed in the weightlessness will be estimated more easily, more inexpensively and more quickly.
This paper proposes gain scheduled H∞ flight control design method, which can satisfy specified stability margin. The parameter changes due to flight conditions are described by LPV (Linear Parameter Varying) model, and gain scheduled controller is designed by LMI (Linear Matrix Inequality) theory. Stability margin requirements can be described as a kind of mixed sensitivity H∞ problem with constant weighting.Therefore, stability margin requirements and another design requirements can be simultaneously satisfied by solving LMI. The effectiveness of this design method is shown by designing longitudinal flight controller for ALFLEX (Automatic Landing FLight EXperiment).
This paper presents a model modification of a high-aspect-ratio aeroelastic wing and active flutter suppression. One of uncertainties in linear models of the aeroelastic wing in the transonic region is included in the unsteady aerodynamic force which is represented by a linear function. To reduce the uncertainty, the linear model of the aeroelasticwing is modified in terms of the unsteady aerodynamic force so that the flutter pressure of the linear model coincides with the one obtained by the wind tunnel test. Using the modified linear model, controllers are then designed by H∞ synthesis for the active flutter suppression. The control performance is evaluated by the wind tunnel test.
Clarification of physical mechanisms on low frequency discharge current oscillation in the 20kHz range is important to improve thruster performance. In this paper, we have investigated the oscillation frequency and the space distribution of oscillation and of main plasma properties (electron temperature and plasma density)under various operating conditions (voltage, magnetic field, channel length, etc.) using a Hall thruster of our design. The results show that the "Ionization oscillation model" could explain qualitatively most of the physical mechanisms for low frequency oscillation phenomenon. The amplitude of oscillation coincides with the reciprocal of the length of ionization zone through comparing the theoretical/numerical analysis which includes this instability against experimental results.
An ion thruster emits a high density plasma plume and the electromagnetic noise radiated from the plasma current may affect the spacecraft instrument. An electromagnetic particle simulation code has been developed to investigate the effects of the electromagnetic interference induced from the ion thruster plume. The electromagnetic waves are excited due to plasma instability caused by strong polarization field inside imperfectly neutralized plume. The strength and frequencies of electromagnetic noise radiated by ion thruster plume depend on the distribution of charged particles inside the plume, static magnetic field strength and its orientation.
Numerical analyses of a one-dimensional flow in an MPD thruster has been done, including nonequilibrium ionization/recombination processes of the propellant gas and velocity slip between charged and neutral particles. Calculated results show that a velocity of charged particles are noticeably different from that of neutral particles downstream of inlet. The velocity of charged particles is higher than the mean velocity calculated by a one-fluid model, and calculated electron temperature of a two-fluid model is much lower than that of the one-fluid model. As a result, more realistic current distribution is obtained using the two-fluid model.