用CNN融合彩色与深度图像，让机器人实时避障
Real-Time Obstacle Avoidance for a Mobile Robot Using CNN-Based Sensor Fusion
- 三套CNN模型从同步图像中端到端学习避障指令
- 最优模型误差仅0.58×10⁻³ rad/s，实时导航成功率100%
- 轻量模型参数少25%、收敛快，适合资源受限场景
障碍物避让是移动机器人在复杂未知环境中有效导航的关键。本研究训练并评估了三种端到端卷积神经网络（CNN），将其部署于差速驱动移动机器人上，利用Intel RealSense D415 RGB-D相机获取的同步彩色与深度图像，实现实时障碍物避让并生成低层转向指令。离线评估显示，NetConEmb模型表现最佳，中位绝对误差（MedAE）仅为0.58×10⁻³ rad/s；相较之下，参数减少约25%、收敛更快的NetEmb架构，其均方根误差（RMSE）为21.68×10⁻³ rad/s，接近NetConEmb的21.42×10⁻³ rad/s。实时导航测试进一步验证了NetConEmb的鲁棒性，在已知与未知环境中均达到100%成功，而NetEmb和NetGated仅能在已知环境中成功导航。
原文摘要 · Abstract (English)
Obstacle avoidance is a critical component of the navigation stack required for mobile robots to operate effectively in complex and unknown environments. In this research, three end-to-end Convolutional Neural Networks (CNNs) were trained and evaluated offline and deployed on a differential-drive mobile robot for real-time obstacle avoidance to generate low-level steering commands from synchronized color and depth images acquired by an Intel RealSense D415 RGB-D camera in diverse environments. Offline evaluation showed that the NetConEmb model achieved the best performance with a notably low MedAE of $0.58 \times 10^{-3}$ rad/s. In comparison, the lighter NetEmb architecture, which reduces the number of trainable parameters by approximately 25\% and converges faster, produced comparable results with an RMSE of $21.68 \times 10^{-3}$ rad/s, close to the $21.42 \times 10^{-3}$ rad/s obtained by NetConEmb. Real-time navigation further confirmed NetConEmb's robustness, achieving a 100\% success rate in both known and unknown environments, while NetEmb and NetGated succeeded only in navigating the known environment.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。