基于神经网络驱动的轻量化自主导航方法及装置

By training and quantizing a deep reinforcement learning network, and utilizing sparse LiDAR data for autonomous navigation of UAVs, the problems of computational complexity and memory consumption of nano-sized UAVs are solved, and autonomous navigation capability under low power consumption is achieved.

CN117906614BActive Publication Date: 2026-07-17BEIJING INFORMATION SCI & TECH UNIV +1

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIJING INFORMATION SCI & TECH UNIV
Filing Date
2024-01-24
Publication Date
2026-07-17

Smart Images

  • Figure CN117906614B_ABST
    Figure CN117906614B_ABST
Patent Text Reader

Abstract

本发明涉及无人飞行器导航技术领域,特别涉及一种基于神经网络驱动的轻量化自主导航方法及装置,其中,方法包括:获取并利用稀疏Lidar数据、无人机相对目标点的极坐标和其前一时刻的线速度和角速度数据训练预设深度强化学习网络;对训练好的深度强化学习网络量化,以将训练好的深度强化学习网络部署在无人机的计算芯片上,得到带有深度强化学习网络的无人机;采集实时稀疏Lidar测距数据、实时极坐标数据和其前一时刻的实时线速度和实时角速度,由带有深度强化学习网络的无人机基于实时数据自主导航至目标位置。由此,解决了现有纳型无人飞行器所采用的神经网络算法往往具有高计算复杂度和大内存占用等问题。
Need to check novelty before this filing date? Find Prior Art