In the tracking of a target using a radar or an imaging sensor, the acceleration of the target is usually modeled as a random vector (or the process noise in the state-space model) with known statistical properties. For a highly maneuvering target, the process noise has a large covariance matrix, and consequently, the estimated state has a large error. This paper proposes an approach that estimates the acceleration from the attitude of the target and uses the estimated acceleration for accurate tracking of the target. An imaging sensor is used for attitude estimation of the target (an aircraft; in this paper) as well as for tracking, without reliance on radar. Our simulation shows that the proposed method can track a maneuvering target much more accurately than traditional Kalman filters.

View full abstract

This paper proposes a new method known as “the line detection method” to detect small pieces of low Earth orbit (LEO) debris. A direction is assumed for the line created by small LEO debris on a CCD image and the values of pixels along that direction are accumulated to improve the signal-to-noise ratio. This method can detect LEO debris that is 30 to 40 times darker than debris that can be detected by usual methods. We tested this method by using the 35-cm telescope and back-illuminated CCD camera at the Mt. Nyukasa Astronomical Observatory. One small piece of LEO debris with a radar cross section of 0.0047 m² was detected. By using this method, the 1-m telescope and back-illuminated wide-field camera at the Bisei Spaceguard Center (BSGC) are expected to be able to detect LEO debris with a size of a few cm. The line detection method will be used to probe the LEO small debris environment and contribute to solving the space debris problem.

View full abstract

Two-dimensional computations of stoichiometric hydrogen-oxygen detonations diluted with nitrogen/argon were performed using a detailed chemical reaction mechanism at initial pressures 0.101 and 0.013 MPa. With increasing channel widths, the relation between the channel widths normalized by half-reaction lengths and the transverse wave strengths defined by pressure ratio across a reflected shock are examined. Various mixture conditions are compared for the maximum channel width where a single transverse wave appears, because numerical cell widths and aspect ratios of the cell are in comparative agreement with previous experimental data. In a mixture diluted with nitrogen at 0.101 MPa, the strong transverse detonation with a transverse wave strength of 1.5 overdrives the transverse wave at 1.3 times faster velocities than the speed of sound in detonation products and makes the maximum channel width larger than those of other conditions. In the presence of strong transverse detonation, acoustic coupling between the transverse wave and the acoustic wave does not apply. The empirical irregularity of the cell with the mixture diluted by nitrogen is interpreted as flexibility of the cell width due to the strong transverse detonation and instability due to out-of-phase acoustic coupling.

View full abstract

Artificial disturbances are introduced to a turbulent boundary layer separating at a convex corner to see how excited vortices control the turbulent separation. The results clearly show that excitation of vortices with appropriate scale can effectively suppress the separation of turbulent boundary layer, not unlike the case of laminar separation. However, strong disturbance growth due to Kelvin-Helmholtz instability, which decisively dominates development of separation bubble for laminar separation, does not occur for turbulent separation. Therefore direct excitation of energetic vortices around the separation point is required for suppression of turbulent separation.

View full abstract

An experiment was carried out to confirm the validity of time series evaluation of supersonic mixing conditions by using the catalytic reaction on a platinum wire. Gaseous hydrogen was injected parallel to a supersonic freestream (M₁≈1.81) from a slit injector, located at a backward-facing step. The time series condition of supersonic mixing was evaluated using a W-type probe with a platinum wire and a reference (nickel) wire. The evaluation was done by simultaneously measuring each electric power supplied by each electric circuit keeping the temperature of the wire constant. Electric power supplied to the Pt wire depended on the catalytic heat release rate (giving hydrogen concentration) and flow convection. Meanwhile electric power supplied to the Ni wire depended on flow convection. The results show that the correlation coefficient between these electric powers increased as mixing developed. Investigations were also conducted for the cases of helium, air, and no secondary injectant to compare with the hydrogen-injectant case. The results indicate that this evaluation method can measure the time-series behavior of the air-hydrogen supersonic mixing layer or coherent motion of turbulence.

View full abstract

Large-Eddy Simulation (LES) is used to simulate the compressible flat plate boundary layer with a Reynolds number up to 5×10⁵. Numerical examples include shock wave/boundary layer interaction and boundary layer transition, aimed at future application to analysis of transonic fan/compressor cascades. The present LES code uses the hybrid compact/WENO scheme for spatial discretization, and the compact diagonalized implicit scheme for time integration. The present code successfully predicted the bypass transition of the subsonic boundary layer. For the supersonic turbulent boundary layer, mean and fluctuation velocity profiles of the attached boundary, as well as the evolution of the friction coefficient and the displacement thickness both upstream and downstream of the separation region are all in good agreement with the experiments. In the simulation of the shock wave/laminar boundary layer interaction, the dependence of the transition upon the strength of the shock wave is reproduced qualitatively. Span-wise disturbance is observed in the separation region, and the disturbance keeps growing when the shock wave is strong. However, it decays at the region between the incident shock wave and the reattachment of the boundary layer, for the weaker incident shock wave. These numerical examples show that LES can predict the behavior of the boundary layer including transition and shock wave interaction, which are poorly managed by the conventional Reynolds-Averaged Navier-Stokes approach. However, more effort is required before achieving quantitative agreement.

View full abstract

We developed an ambiguity resolution algorithm for attitude determination using multiple GPS antennas. It is focused on reducing computational loads using a geometrical approach. By defining a new decision variable for stochastic verification, overall computational loads are effectively reduced with respect to the chi-square model. This paper also uses the first base vector estimated as line-of-sight vector to obtain a second base vector. The additional 1-D search space is sufficient to resolve cycle ambiguity of the second base vector for complete attitude determination. Consequently it is possible to resolve the multi-baseline cycle ambiguity using a single baseline algorithm load.

View full abstract

This paper presents a numerical study on surface melting pattern formation due to aerodynamic heating. When aerodynamic heating is severe enough to melt the surface of a flying body, the interaction between the melted layer and the external flow creates surface patterns. The present study succeeds in reproducing surface melting patterns, and results show the same trend as predicted by theory and experiment. The instability mechanism is mainly governed by surface friction and surface pressure fluctuations. These two factors act as driving forces on the surface. The surface pattern formation is characterized by the amplification factor and the wavelength factor. The results show the same trend as the experiments and the validity of the present analysis has been demonstrated.

View full abstract