Space debris detection with fast grid-based learning

Xiang, Yaobing; Xi, Jiangbo; Cong, Ming; Ye, Yun; Ren, Chaofeng; Han, Ling

doi:10.1109/iicspi51290.2020.9332372

Cited by 13 publications

(8 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, Reinforcement learning (RL) has been explored in SSA [28, 29]. Reinforcement learning is a type of learning that allows an agent to learn through interactions with an environment, receiving rewards or punishments for actions taken.…”

Section: Related Workmentioning

confidence: 99%

Deep learning‐based space debris detection for space situational awareness: A feasibility study applied to the radar processing

Massimi,

Ferrara,

Petrucci

et al. 2024

IET Radar Sonar & Navi

View full text Add to dashboard Cite

The increasing number of space objects (SO), debris, and constellation of satellites in Low Earth Orbit poses a significant threat to the sustainability and safety of space operations, which must be carefully and efficiently addressed to avoid mutual collisions. The space situational awareness is currently addressed by an ensemble of radar and radio‐telescopes that detect and track SO. However, a large part of space debris is composed of very small and tiny metallic objects, very difficult to detect. The authors demonstrate the benefits of using deep learning (DL) architectures for small space object detection by radar observations. TIRA radio telescope has been simulated to generate range‐Doppler maps, then used as inputs for object detection exploiting You‐Only‐Look‐Once (YOLO) frameworks. The results demonstrate that the object detection by using YOLO algorithms outperform conventional target detection approaches, thus indicating the potential benefits of using DL techniques for space surveillance applications.

show abstract

Section: Related Workmentioning

confidence: 99%

Deep learning‐based space debris detection for space situational awareness: A feasibility study applied to the radar processing

Massimi,

Ferrara,

Petrucci

et al. 2024

IET Radar Sonar & Navi

View full text Add to dashboard Cite

show abstract

“…3 (right), we can see the impact of debris hit in ISS. In [8,20], the authors have mapped the debris and detect them (Micrometeoroids and Orbital Debris). These objects can cause sandblasting, especially to spacecraft appendages and optics that are difficult to be protected by any shield [1].…”

Section: Satellite Applicationmentioning

confidence: 99%

Modeling and Quickest Detection of a Rapidly Approaching Object

Tim¹,

Banerjee²,

Mishra³

2023

Preprint

View full text Add to dashboard Cite

The problem of detecting the presence of a signal that can lead to a disaster is studied. A decision-maker collects data sequentially over time. At some point in time, called the change point, the distribution of data changes. This change in distribution could be due to an event or a sudden arrival of an enemy object. If not detected quickly, this change has the potential to cause a major disaster. In space and military applications, the values of the measurements can stochastically grow with time as the enemy object moves closer to the target. A new class of stochastic processes, called exploding processes, is introduced to model stochastically growing data. An algorithm is proposed and shown to be asymptotically optimal as the mean time to a false alarm goes to infinity.

show abstract

“…Table 3 shows the number of overlapping cycles and calculation time for each calculation process. The design principle of overlapping cycles between processes is to ensure that the input used in the next level process calculation is the output completed by the previous level process calculation, which is searched and filtered in D(n) ∈ [3,7] (k = 1). Parallel computing processes such as s1.1 and s1.2, due to simultaneous computation, the parallelism is based on the longest module time, and the selection of the parallelism for the second longest module is based on the calculation time not exceeding the longest module calculation time.…”

Section: Parallelism Parameter Calculationmentioning

confidence: 99%

“…The key of this process lies in the real‐time calculation of target detection and tracking. After long‐term development, a mature framework has been formed for interframe sequence detection and tracking 7 . In the tracking process, the adoption of parallel computing architecture to complete the Kalman filter calculation in real time is the key to the effective completion of the whole framework on‐board 8 …”

Section: Introductionmentioning

confidence: 99%

Parallel accelerated computing architecture for dim target tracking on‐board

Yu,

Huang,

et al. 2023

Computational Intelligence

View full text Add to dashboard Cite

The real‐time tracking process of dim targets in space is mainly achieved through the correlation and prediction of dots after the detection and calculation process. The on‐board calculation of the tracking needs to be completed in milliseconds, and it needs to reach the microsecond level at high frame rates. For real‐time tracking of dim targets in space, it is necessary to achieve universal tracking calculation acceleration in response to different space regions and complex backgrounds, which poses high requirements for engineering implementation architecture. This paper designs a Kalman filter calculation based on digital logic parallel acceleration architecture for real‐time solution of dim target tracking on‐board. A unified architecture of Vector Processing Element (VPE) was established for the calculation of Kalman filtering matrix, and an array computing structure based on VPE was designed to decompose the entire filtering process and form a parallel pipelined data stream. The prediction errors under different fixed point bit widths were analyzed and deduced, and the guidance methods for selecting the optimal bit width based on the statistical results were provided. The entire design was engineered based on Xilinx's XC7K325T, resulting in an energy efficiency improvement compared to previous designs. The single iteration calculation time does not exceed 0.7 microseconds, which can meet the current high frame rate target tracking requirements. The effectiveness of this design has been verified through simulation of random trajectory data, which is consistent with the theoretical calculation error.

show abstract

Space debris detection with fast grid-based learning

Cited by 13 publications

References 13 publications

Deep learning‐based space debris detection for space situational awareness: A feasibility study applied to the radar processing

Deep learning‐based space debris detection for space situational awareness: A feasibility study applied to the radar processing

Modeling and Quickest Detection of a Rapidly Approaching Object

Parallel accelerated computing architecture for dim target tracking on‐board

Contact Info

Product

Resources

About