Research on UAV Perception Systems for Dynamic Indoor Environments: PointNet++ Optimization with Hierarchical Contextual Attention

Yuhan Jiang

doi:10.47297/taposatWSP2633-456915.20250605

PDF
导出
分享
收藏
专辑

Research on UAV Perception Systems for Dynamic Indoor Environments: PointNet++ Optimization with Hierarchical Contextual Attention

6卷第5期

浏览量: 0 Downloads: 0

Research on UAV Perception Systems for Dynamic Indoor Environments: PointNet++ Optimization with Hierarchical Contextual Attention
Research on UAV Perception Systems for Dynamic Indoor Environments: PointNet++ Optimization with Hierarchical Contextual Attention
Department of Mechanical Engineering, City University of Hong Kong, Hong Kong, 999077, China
2025年6卷第5期页码：68-73
纸质出版日期：2025-05-19
DOI：10.47297/taposatWSP2633-456915.20250605
Accepted：

Scan QR Code

(2025). Research on UAV Perception Systems for Dynamic Indoor Environments: PointNet++ Optimization with Hierarchical Contextual Attention. 科技理论与实践（英文版）, 6(5), 68-73.

Jiang Yuhan.Research on UAV Perception Systems for Dynamic Indoor Environments: PointNet++ Optimization with Hierarchical Contextual Attention[J].Theory and Practice of Science and Technology,2025,06(05):68-73.
(2025). Research on UAV Perception Systems for Dynamic Indoor Environments: PointNet++ Optimization with Hierarchical Contextual Attention. 科技理论与实践（英文版）, 6(5), 68-73. DOI： 10.47297/taposatWSP2633-456915.20250605.

Jiang Yuhan.Research on UAV Perception Systems for Dynamic Indoor Environments: PointNet++ Optimization with Hierarchical Contextual Attention[J].Theory and Practice of Science and Technology,2025,06(05):68-73. DOI： 10.47297/taposatWSP2633-456915.20250605.

摘要

Abstract

This research addresses the challenges of dynamic perception systems for UAVs in indoor environments by proposing an optimization method for PointNet++ that integrates hierarchical contextual attention mechanisms. To address the issue of imprecise boundary feature extraction in traditional LiDAR point cloud processing for dynamic object detection and tracking

we designed a Hierarchical Contextual Attention Feature Propagation (HCAFP) module. This module establishes weighted connections between high-level semantic features and low-level geometric features

significantly enhancing the model's perception capabilities for object part boundaries. Experimental results show that the enhanced model achieved an instance average IoU of 86.7% on the ShapeNetCore dataset

an improvement of 1.3 percentage points over the original PointNet++ MSG architecture. Additionally

we proposed a series of model optimization strategies

including model compression and quantization

lightweight model structure design

and hierarchical sparse attention mechanisms

enabling the improved model to be efficiently deployed on UAV platforms with limited computational resources while maintaining competitive performance (85.3% instance average IoU). This research provides a new technical approach for UAV autonomous navigation and dynamic object perception in indoor environments

with significant implications for UAV applications in complex dynamic indoor settings.

关键词

Keywords

references

KATONA K, NEAMAH H A, KORONDI P. Obstacle Avoidance and Path Planning Methods for Autonomous Navigation of Mobile Robot[J]. Sensors (14248220), 2024, 24(11).

FAN, BANGKUI, LI, et al. Review on the Technological Development and Application of UAV Systems[J]. Chinese Journal of Electronics, 2020, v.29(02): 5-13.

DU W, BELTRAME G. LiDAR-based Real-Time Object Detection and Tracking in Dynamic Environments[J]. 2024.

DE CARVALHO M V L, SIMONI R, YOSHIOKA L. Autonomous Navigation and Collision Avoidance for Mobile Robots: Classification and Review[J]. 2024.

KLANAR G, SEDER M, BLAI S. Advanced Sensors Technologies Applied in Mobile Robot[J]. Sensors (14248220), 2023, 23(6).

REN Y, ZHU F, LU G, et al. Safety-assured high-speed navigation for MAVs[J]. Science Robotics, 2025, 10(98).

QI C R, YI L, SU H, et al. PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space[J]. 2017.

FAHAD SAFDAR M, SHARIF H, REHMAN F, et al. A Review of Recent Advances in Deep Learning for Object Detection[C/OL]//2022 3rd International Conference on Innovations in Computer Science & Software Engineering (ICONICS). 2022: 1-6. DOI:10.1109/ICONICS56716.2022.10100486.

TANAKA H, KUNIN D, YAMINS D L K, et al. Pruning neural networks without any data by iteratively conserving synaptic flow[J]. 2020.

HOLMBERG M, KARLSSON O, TULLDAHL M. Lidar Positioning for Indoor Precision Navigation[C/OL]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). 2022: 358-367. DOI:10.1109/CVPRW56347.2022.00051.

JACOB B, KLIGYS S, CHEN B, et al. Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2018.

HINTON G E, VINYALS O, DEAN J. Distilling the Knowledge in a Neural Network[J/OL]. ArXiv, 2015, abs/1503.02531. https://api.semanticscholar.org/CorpusID:7200347.

文章被引用时，请邮件提醒。

Submit

No data

Related Author

No data

Related Institution

No data

⁰