EdgeLPR: On the Deep Neural Network trade-off between Precision and Performance in LiDAR Place Recognition

TL;DR

This paper investigates the trade-offs between accuracy, robustness, and efficiency in LiDAR-based place recognition for EdgeAI, focusing on lightweight neural networks and quantization effects.

Contribution

It benchmarks various lightweight architectures using Bird's Eye View representations and analyzes quantization impacts for resource-constrained deployment.

Findings

FP16 quantization maintains accuracy similar to FP32 with reduced cost.

INT8 quantization causes architecture-dependent degradation in performance.

Bird's Eye View representations enable effective lightweight LiDAR place recognition.

Abstract

Place recognition is essential for long-term autonomous navigation, enabling loop closure and consistent mapping. Although deep learning has improved performance, deploying such models on resource-constrained platforms remains challenging. This work explores efficient LiDAR-based place recognition for EdgeAI by leveraging Bird's Eye View representations to enable lightweight image-based networks. We benchmark representative architectures without aggregation heads using a unified descriptor scheme based on global pooling and linear projection, and evaluate performance under FP32, FP16, and INT8 quantization. Experiments reveal trade-offs between accuracy, robustness, and efficiency: FP16 matches FP32 with lower cost, while INT8 introduces architecture-dependent degradation. Overall, the presented results are a strong basis for future research on 'use-case'-aware quantisation of Neural Networks for Edge deployment.