Naoya Ozaki

Dario Izzo, Daniel Hennes, Marcus Märtens et al.

We consider the interplanetary trajectory design problem posed by the 8th edition of the Global Trajectory Optimization Competition and present the end-to-end strategy developed by the team ACT-ISAS (a collaboration between the European Space Agency's Advanced Concepts Team and JAXA's Institute of Space and Astronautical Science). The resulting interplanetary trajectory won 1st place in the competition, achieving a final mission value of $J=146.33$ [Mkm]. Several new algorithms were developed in this context but have an interest that go beyond the particular problem considered, thus, they are discussed in some detail. These include the Moon-targeting technique, allowing one to target a Moon encounter from a low Earth orbit; the 1-$k$ and 2-$k$ fly-by targeting techniques, enabling one to design resonant fly-bys while ensuring a targeted future formation plane% is acquired at some point after the manoeuvre ; the distributed low-thrust targeting technique, admitting one to control the spacecraft formation plane at 1,000,000 [km]; and the low-thrust optimization technique, permitting one to enforce the formation plane's orientations as path constraints.

Masahiro Fujiwara, Naoya Ozaki

Designing spacecraft trajectories remains challenging in the presence of stochastic effects such as maneuver execution errors and observation uncertainties. Although covariance control and belief-space planning provide useful tools for designing robust control policies and information-aware trajectories under uncertainty, practical methods remain limited for partially observable trajectory optimization problems in which trajectory design, orbit determination, and correction maneuver planning are tightly coupled. This paper presents a stochastic differential dynamic programming algorithm for such coupled problems. The proposed method optimizes the nominal control sequence and feedback gains subject to belief dynamics and general mission constraints, explicitly accounting for the dependence of covariance propagation on the nominal trajectory without relying on the separation principle. Numerical examples demonstrate that the proposed algorithm produces navigation-aware and uncertainty-robust solutions across a range of dynamical systems, observation models, and uncertainty levels. In particular, the circular restricted three-body problem shows that the proposed method can exploit the coupling between trajectory design and orbit determination to obtain navigation-aware solutions with substantially lower fuel consumption than those from deterministic local optimization starting from the same initial guess.

Masato Shibukawa, Fumi Yoshida, Toshifumi Yanagisawa et al.

One of the greatest challenges for detecting moving objects in the solar system from wide-field survey data is determining whether a signal indicates a true object or is due to some other source, like noise. Object verification has relied heavily on human eyes, which usually results in significant labor costs. In order to address this limitation and reduce the reliance on manual intervention, we propose a multi-input convolutional neural network integrated with a convolutional block attention module. This method is specifically tailored to enhance the moving object detection system that we have developed and used previously. The current method introduces two innovations. This first one is a multi-input architecture that processes multiple stacked images simultaneously. The second is the incorporation of the convolutional block attention module which enables the model to focus on essential features in both spatial and channel dimensions. These advancements facilitate efficient learning from multiple inputs, leading to more robust detection of moving objects. The performance of the model is evaluated on a dataset consisting of approximately 2,000 observational images. We achieved an accuracy of nearly 99% with AUC (an Area Under the Curve) of >0.99. These metrics indicate that the proposed model achieves excellent classification performance. By adjusting the threshold for object detection, the new model reduces the human workload by more than 99% compared to manual verification.

Masato Shibukawa, Naoya Ozaki, Maximilien Berthet

Particle methods play an important role in computational fluid dynamics, but they are among the most difficult to implement and solve. The most common method is smoothed particle hydrodynamics, which is suitable for problem settings that involve large deformations, such as tsunamis and dam breaking. However, the calculation can become unstable depending on the distribution of particles. In contrast, the neural particle method has high computational stability for various particle distributions is a machine learning method that approximates velocity and pressure in a spatial domain using neural networks. The neural particle method has been extended to viscous flows, but until now it has been limited to incompressible flows. In this paper, we propose the compressible neural particle method, which is a new feed-forward neural network-based method that extends the original neural particle method to model compressible viscous fluid flows. The proposed method uses neural networks to calculate the velocity and pressure of fluid particles at the next time step, and the Tait equation to calculate the density to handle the compressibility. The loss function is composed of the governing equations of compressible flow and the boundary conditions, which are free surface and solid boundary conditions. We demonstrate that the proposed method can accurately solve the compressible viscous fluid flow, a problem that was difficult to solve with the smoothed particle hydrodynamics method, by applying it to a dam breaking problem.

Takayuki Hosonuma, Takeshi Miyabara, Naoya Ozaki et al.

DESTINY+ is an upcoming JAXA Epsilon medium-class mission to flyby multiple asteroids including Phaethon. As an asteroid flyby observation instrument, a telescope mechanically capable of single-axis rotation, named TCAP, is mounted on the spacecraft to track and observe the target asteroids during flyby. As in past flyby missions utilizing rotating telescopes, TCAP is also used as a navigation camera for autonomous optical navigation during the closest-approach phase. To mitigate the degradation of the navigation accuracy, past missions performed calibration of the navigation camera's alignment before starting optical navigation. However, such calibration requires significant operational time to complete and imposes constraints on the operation sequence. From the above background, the DESTINY+ team has studied the possibility of reducing operational costs by allowing TCAP alignment errors to remain. This paper describes an autonomous optical navigation algorithm robust to the misalignment of rotating telescopes, proposed in this context. In the proposed method, the misalignment of the telescope is estimated simultaneously with the spacecraft's orbit relative to the flyby target. To deal with the nonlinearity between the misalignment and the observation value, the proposed method utilizes the unscented Kalman filter, instead of the extended Kalman filter widely used in past studies. The proposed method was evaluated with numerical simulations on a PC and with hardware-in-the-loop simulation, taking the Phaethon flyby in the DESTINY+ mission as an example. The validation results suggest that the proposed method can mitigate the misalignment-induced degradation of the optical navigation accuracy with reasonable computational costs suited for onboard computers.

Akira Hatakeyama, Shota Ito, Toshihiko Yanase et al.

Two-point boundary value problems for cislunar trajectories present significant challenges in circler restricted three body problem, making traditional analytical methods like Lambert's problem inapplicable. This study proposes a novel approach using a prefixed patch time series Transformer model that automates the solution of two-point boundary value problems from lunar flyby to arbitrary terminal conditions. Using prefix tokens of terminal conditions in our deep generative model enables solving boundary value problems in three-body dynamics. The training dataset consists of trajectories obtained through forward propagation rather than solving boundary value problems directly. The model demonstrates potential practical utility for preliminary trajectory design in cislunar mission scenarios.

Naoya Ozaki, Kanta Yanagida, Takuya Chikazawa et al.

Asteroid exploration has been attracting more attention in recent years. Nevertheless, we have just visited tens of asteroids while we have discovered more than one million bodies. As our current observation and knowledge should be biased, it is essential to explore multiple asteroids directly to better understand the remains of planetary building materials. One of the mission design solutions is utilizing asteroid flyby cycler trajectories with multiple Earth gravity assists. An asteroid flyby cycler trajectory design problem is a subclass of global trajectory optimization problems with multiple flybys, involving a trajectory optimization problem for a given flyby sequence and a combinatorial optimization problem to decide the sequence of the flybys. As the number of flyby bodies grows, the computation time of this optimization problem expands maliciously. This paper presents a new method to design asteroid flyby cycler trajectories utilizing a surrogate model constructed by deep neural networks approximating trajectory optimization results. Since one of the bottlenecks of machine learning approaches is the computation time to generate massive trajectory databases, we propose an efficient database generation strategy by introducing pseudo-asteroids satisfying the Karush-Kuhn-Tucker conditions. The numerical result applied to JAXA's DESTINY+ mission shows that the proposed method is practically applicable to space mission design and can significantly reduce the computational time for searching asteroid flyby sequences.