Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper addresses the challenges of airborne surveillance from nanosatellites by integrating on-board inference with generative data augmentation to enhance aircraft detection capabilities. By employing a 6U CubeSat with a low-power edge tensor accelerator and a diffusion model for synthetic image generation, the authors significantly improve the representation of minority aircraft classes in the dataset. The proposed workflow not only boosts the global mean average precision from 77.9% to 82.2% but also enhances the F1 score for the minority class from 0.683 to 0.811, demonstrating its effectiveness as a real-time decision-support tool for autonomous surveillance.
Generative data augmentation can dramatically enhance aircraft detection in nanosatellite surveillance, boosting minority class performance by over 18% in F1 score.
Airborne surveillance from low Earth orbit is hindered by two interconnected bottlenecks: nanosatellites have a limited downlink budget, yet the conventional approach still transmits terabytes of raw imagery to the ground for processing, and open satellite datasets for aircraft are scarce and severely class-imbalanced. These limitations either delay timely decision-making or prevent standard detectors from learning robust representations of rare aircraft classes. In this paper, a workflow that combines on-board inference with generative data augmentation is proposed to address both limitations jointly. Inference is executed on a 6U CubeSat equipped with a low-power edge tensor accelerator, while a diffusion model fine-tuned through low-rank adaptation generates synthetic minority-class imagery. This synthetic output is automatically annotated, pseudo-labelled, by an intermediate detector and merged with classically augmented samples. The results show that the balanced dataset increases global mean average precision from 77.9% to 82.2%, with the minority class rising from F1=0.683 to F1=0.811, and that the quantised detector fits the on-chip memory and projects 25-30 frames per second on orbit. This approach contrasts with the conventional bent-pipe architecture, in which the satellite acts as a passive data collector. Therefore, the computational tests support the proposed workflow as a decision-support tool for real-time, autonomous airborne surveillance from nanosatellites.