Real-Time Spacecraft Pose Estimation Using Mixed-Precision Quantized Neural Network on COTS Reconfigurable MPSoC

TL;DR

This paper introduces a real-time, energy-efficient spacecraft pose estimation method using a mixed-precision quantized neural network on FPGA, achieving significant speed and energy improvements with an open-source implementation.

Contribution

It presents the first real-time, open-source FPGA-based spacecraft pose estimation system with a novel co-design and evaluation methodology for quantization.

Findings

7.7 times faster than previous methods

19.5 times more energy-efficient

First open-source real-time implementation

Abstract

This article presents a pioneering approach to real-time spacecraft pose estimation, utilizing a mixed-precision quantized neural network implemented on the FPGA components of a commercially available Xilinx MPSoC, renowned for its suitability in space applications. Our co-design methodology includes a novel evaluation technique for assessing the layer-wise neural network sensitivity to quantization, facilitating an optimal balance between accuracy, latency, and FPGA resource utilization. Utilizing the FINN library, we developed a bespoke FPGA dataflow accelerator that integrates on-chip weights and activation functions to minimize latency and energy consumption. Our implementation is 7.7 times faster and 19.5 times more energy-efficient than the best-reported values in the existing spacecraft pose estimation literature. Furthermore, our contribution includes the first real-time, open-source implementation of such algorithms, marking a significant advancement in making efficient spacecraft pose estimation algorithms widely accessible. The source code is available at https://github.com/possoj/FPGA-SpacePose.