使用神经网络进行3D空中声呐的语义路标检测与分类
Semantic Landmark Detection & Classification Using Neural Networks For 3D In-Air Sonar
AI总结：
提出一种基于卷积神经网络和空中3D声呐的方法，通过耳蜗图检测分类十种反射器路标并预测朝向角，在室内测试中达97.3%准确率且角度RMSE低于10度，有效提升了SLAM与自主导航的鲁棒性。
摘要：
在传统传感方式难以应对的挑战性环境中，空中声呐对光学干扰具有鲁棒性。在这些环境中放置先验已知的路标可以消除自主移动系统（如同步定位与建图（SLAM）和自主导航）中的累积误差。我们提出了一种新颖的方法，使用卷积神经网络通过空中3D声呐检测和分类十种具有不同半径的反射器路标。此外，该网络还能预测检测到的路标的朝向角。该神经网络在耳蜗图上进行训练，耳蜗图表示传感器在时频域中接收到的回波。在杂乱的室内环境中的实验结果显示出有前景的性能。该CNN在测试数据集上实现了97.3%的分类准确率，能够准确检测路标的存在与缺失。此外，该网络预测路标朝向角的RMSE低于10度，增强了其在SLAM和自主导航应用中的实用性。这一进展提高了自主系统在挑战性环境中的鲁棒性和准确性。
英文摘要：
In challenging environments where traditional sensing modalities struggle, in-air sonar offers resilience to optical interference. Placing a priori known landmarks in these environments can eliminate accumulated errors in autonomous mobile systems such as Simultaneous Localization and Mapping (SLAM) and autonomous navigation. We present a novel approach using a convolutional neural network to detect and classify ten different reflector landmarks with varying radii using in-air 3D sonar. Additionally, the network predicts the orientation angle of the detected landmarks. The neural network is trained on cochleograms, representing echoes received by the sensor in a time-frequency domain. Experimental results in cluttered indoor settings show promising performance. The CNN achieves a 97.3% classification accuracy on the test dataset, accurately detecting both the presence and absence of landmarks. Moreover, the network predicts landmark orientation angles with an RMSE lower than 10 degrees, enhancing the utility in SLAM and autonomous navigation applications. This advancement improves the robustness and accuracy of autonomous systems in challenging environments.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
