Optical sensor systems are essential for space target detection. However, previous studies have prioritized detection accuracy over model efficiency, limiting their deployment on resource-constrained sensors. To address this issue, we propose the lightweight space target real-time detection transformer (STar-DETR), which achieves a balance between model efficiency and detection accuracy. First, the improved MobileNetv4 (IMNv4) backbone network is developed to significantly reduce the model’s parameters and computational complexity. Second, group shuffle convolution (GSConv) is incorporated into the efficient hybrid encoder, which reduces convolution parameters while facilitating information exchange between channels. Subsequently, the dynamic depthwise shuffle transformer (DDST) feature fusion module is introduced to emphasize the trajectory formed by space target exposure. Finally, the minimum points distance scylla intersection over union (MPDSIoU) loss function is developed to enhance regression accuracy and expedite model convergence. A space target dataset is constructed, integrating offline and online data augmentation techniques to improve robustness under diverse sensing conditions. The proposed STar-DETR model achieves an $A{P}_{0.5:0.95}$ of 89.9%, successfully detecting dim and discontinuous streak space targets. Its parameter count and computational complexity are reduced by 64.8% and 41.8%, respectively, highlighting its lightweight design and providing a valuable reference for space target detection in resource-constrained optical sensors.

As the number of satellites and amount of space debris in Low-Earth orbit (LEO) increase, high-precision orbit determination is crucial for ensuring the safe operation of spacecraft and maintaining space situational awareness. However, ground-based optical observations are constrained by limited arc-segment angular data and dynamic noise interference, and the traditional Extended Kalman Filter (EKF) struggles to meet the accuracy and robustness requirements in complex orbital environments. To address these challenges, this paper proposes a Bayesian Adaptive Extended Kalman Filter (BAEKF), which synergistically optimizes track determination through dynamic noise covariance adjustment and Bayesian a posteriori probability correction. Experiments demonstrate that the average root mean square error (RMSE) of BAEKF is reduced by 34.7% compared to the traditional EKF, effectively addressing EKF’s accuracy and stability issues in nonlinear systems. The RMSE values of UKF, RBFNN, and GPR also show improvement, providing a reliable solution for high-precision orbital determination using optical observation.