Reconfigurability in versatile systems of modular robots is achieved by appropriately actuating individual modular units. Optimizing the distribution of active and passive modules in modular architecture can significantly reduce both cost and energy of a reconfiguration task. This paper presents a methodology for planning this distribution in modular robots, resulting in a minimum number of active modules that guarantees the capability to reconfigure. We discuss the optimal distribution problem in layout-based and target-based planning schemes such that modular robots can instantly respond to reconfiguration commands with either an initial planar layout or a target configuration as input. We propose heuristic algorithms as solutions for the different scenarios, which we demonstrate by applying them to Mori, a modular origami robot, in simulation. The results show that our algorithms yield high-quality distribution schemes in reduced time, and are thus viable for real-time applications in modular robotic systems.

Designing spacecraft involves a careful equilibrium to avoid overengineering or underdesigning, which underscores the importance of employing thermal uncertainty analysis. A key part of this analysis is modeling thermal conditions, but this is often a computationally heavy process. This is largely because ray-tracing calculations require determining the external heat flux of solar radiation across different operating conditions. Ray emission varies across conditions, which can lead to inefficient resource use in uncertainty calculations. Our study aims to address this by introducing a new approach to calculating the external heat flux of solar radiation that is better suited for uncertainty analysis than previous approaches. Our formula only requires ray tracing to be performed for one condition rather than for every condition. The other conditions are handled by simple matrix budgeting, negating the need for complicated ray tracing. In the aforementioned analytical procedure, certain matrices demonstrate sparsity properties. By exploiting this characteristic, optimization computations can be executed by utilizing sparse matrix algorithms. We tested this new formula, which we call the external heat flux expansion (EHFE) formula, on a specific spacecraft and compared the results with those obtained using the traditional method. Our findings suggest that the EHFE formula is ideal for calculating uncertainty. It significantly improves computational efficiency while maintaining accuracy. The formula is also user-adjustable, allowing the accuracy of uncertainty calculation results of the external heat flux of solar radiation to be fine-tuned by changing the value of the cutoff factor. This work establishes an essential theoretical framework pivotal to addressing inherent uncertainties in the thermal design of upcoming deep-space exploration spacecraft, solar observatory satellites, and space solar power stations.

This paper proposes a new singularity direction escape steering law for control moment gyros (CMG) using singular value decomposition (SVD). The proposed steering law differs from the classical singular direction avoidance (SDA) steering law in two respects. First, torque errors are introduced in the direction of the smallest two singular values through modification of the diagonal matrix. This is done by tuning parameters related to the magnitude of torque error so that both singularity avoidance and escape performance can be achieved. Second, instead of using a conventional singularity measure, new output torque metrics of the CMG cluster along the torque command direction are introduced and adopted in this steering law, so as to reduce unnecessary torque error for certain near-singular circumstances. Several numerical simulations are provided at last, and comparisons with existing steering laws are presented to demonstrate the superior performance of the proposed method.

With the continuous deepening of Mars exploration missions, the Mars helicopter has become a key platform for acquiring high-resolution near-ground imagery. However, accurate semantic segmentation of the Martian surface remains challenging due to complex terrain morphology, sandstorm interference, and the limited onboard computational resources that restrict real-time processing. Existing models either introduce high computational overhead unsuitable for deployment on Mars aerial platforms or fail to jointly capture fine-grained local texture and global contextual structure information. To address these limitations, we propose LisseMars, a lightweight semantic segmentation network designed for efficient onboard perception. The model integrates a Window Movable Attention (WMA) module for enhanced global context extraction and a multi-convolutional feedforward module (CFFN) to strengthen local detail representation. A Dynamic Polygon Convolution (DPC) module is further introduced to improve segmentation performance on geometrically heterogeneous objects, while a Group Fusion Module (GFM) enables effective multi-scale semantic integration. Extensive experiments are conducted on both real Tianwen-1 Mars helicopter imagery and synthetic datasets. The results show that our method achieved a mean IoU of 78.56% with only 0.12 MB of model parameters, validating the effectiveness of the proposed framework. The real-time performance of proposed method on edge device deployment further demonstrate potential application for real Mars airborne missions.

Thermal uncertainty analysis of spacecraft is an important method to avoid overdesign and underdesign problems. In the context of uncertainty analysis, thermal models representing multiple operating conditions must be invoked repeatedly, leading to substantial computational costs. The ray tracing calculation of Earth infrared and albedo radiation heat flux is an important reason for the slow calculation speed. As the rays emitted during external heat flux calculations under different operating conditions are independent and unconnected, the rays produced across various conditions are effectively wasted. In this study, the external heat flow equation is thoroughly expanded and the derived factors are clustered and analyzed to develop a novel formula for calculating external heat flow. When this formula is employed to compute the uncertain external heat flux, only one condition necessitates ray tracing, while the remaining conditions utilize simple matrix operations in place of complex ray tracing. Within the aforementioned procedure, certain matrices demonstrate sparse characteristics. The optimization calculations for these matrices can, therefore, benefit from the application of sparse matrix optimization algorithms. Using a spacecraft as an example, the uncertain external heat flux calculation outcomes of the new and traditional formulas are compared and assessed. The findings reveal that the new formula is highly suitable for estimating uncertain Earth radiation heat flow, with a marked improvement in efficiency. The accuracy is essentially equivalent to that of the traditional formula and the calculation precision can be dynamically adjusted to meet user requirements. The methodology can be further generalized to assess the uncertainties associated with radiative external heat fluxes for other celestial bodies within the solar system. This offers a valuable theoretical framework for addressing the uncertainties in the thermal design of deep space exploration vehicles.

With the development of convolutional neural networks (CNNs), the application of learning-based methods in aerospace attracts much attention. Compared with sufficient datasets in the generic scene, most of the datasets in aerospace lack diversity in model structure, which leads to the low generalization ability in on-orbital tasks, such as satellite detection. In this work, this drawback is relieved by a synthetic dataset whose structure diversity is extended by 21 models in Yale-CMU-Berkely Object and Model set (YCB dataset) and 2 realistic satellites. This dataset is rendered in a low-earth-orbit scene and is utilized in target detection tasks. Experiment shows unknown satellites can be detected in our synthetic images. The generalization is also evaluated by combining our dataset with other public satellite datasets. The afore-mentioned experiments are implemented by YOLOv5s and YOLOv5m, we also propose a light-weighted refinement pipeline that improves the performance for an average of 2.57% in F1-score in the generalization experiments.