关于为从事电力塔检查任务的视觉引导无人机设计基于深度学习的控制算法
Guillaume Maitre1, Dimitri Martinot2, Elio Tuci1
1Faculty of Computer Science, University of Namur, Namur, Belgium.
Frontiers in robotics and AI
|May 13, 2024
概括
训练卷积神经网络与合成和真实图像的混合数据集改善了无人机 (UAV) 检查. 这种方法增强了用于自主电力塔检查的图像细分,优于在单个数据类型上训练的模型.
科学领域:
- 计算机视觉 计算机视觉
- 机器人技术 机器人技术 机器人技术
- 人工智能的人工智能
背景情况:
- 自主无人飞行器 (UAV) 需要强大的视觉指导来执行诸如电力塔检查等任务.
- 对于无人机来说,图像细分是具有挑战性的,因为塔楼结构和复杂的背景不同.
- 用于培训的手动数据注释是劳动密集型和昂贵的.
研究的目的:
- 设计和评估卷积神经网络 (CNN),用于在电力塔检查中对无人机进行视觉指导.
- 为了比较在合成,物理世界和混合图像数据集上训练的CNN的性能.
- 评估摄影计在生成合成训练数据方面的有效性.
主要方法:
- 使用基于注意力的U-NET架构进行图像细分.
- 使用摄影计和模拟无人机环境生成合成图像.
- 进行了一项使用合成,物理世界和混合数据集进行网络培训的比较研究.
- 使用多个图像细分指标评估网络性能.
主要成果:
- 在混合数据集上训练的CNN显著超过那些仅在合成或物理世界数据上训练的人.
- 混合数据集培训在各种图像细分评估指标中显示出卓越的性能.
- 摄影仪表在创建无人机检查任务的有价值的合成数据集方面被证明是有效的.
结论:
- 结合合成和物理世界的图像的混合方法为训练无人机检查网络提供了成本效益和性能之间的最佳平衡.
- 该研究验证了使用光度计生成数据来自动化精确无人机移动用于基础设施检查的潜力.
相关概念视频
PI Controller: Design
250
Proportional Integral (PI) controllers are a fundamental component in modern control systems, widely used to enhance performance and mitigate steady-state errors. They are particularly effective in applications such as automatic brightness adjustment on smartphones, where they excel at mitigating steady-state errors for step-function inputs. Unlike PD controllers, which require time-varying errors to function optimally, PI controllers leverage their integral component to address residual...
250
PD Controller: Design
222
In automotive engineering, car suspension systems often employ Proportional Derivative (PD) controllers to enhance performance. PD controllers are utilized to adjust the damping force in response to road conditions. A controller, acting as an amplifier with a constant gain, demonstrates proportional control, with output directly mirroring input.
Designing a continuous-data controller requires selecting and linking components like adders and integrators, which are fundamental in Proportional,...
Designing a continuous-data controller requires selecting and linking components like adders and integrators, which are fundamental in Proportional,...
222


