论文标题
SE（3）上具有恒定加速运动的雷达进气测前和极性测量模型
Radar Odometry on SE(3) with Constant Acceleration Motion Prior and Polar Measurement Model
论文作者
论文摘要
本文介绍了$ SE（3）$上的雷达进程方法的方法，该方法利用了恒定的加速运动。运动先验被整合到滑动窗口优化方案中。我们使用Magnus扩展来准确整合运动，同时保持实时性能。此外，我们采用极地测量模型来更好地表示雷达检测不确定性。使用原型高分辨率雷达传感器的大型现实世界数据集评估我们的估计器。新运动先验和测量模型显着提高了相对于先前工作的恒定速度运动和笛卡尔测量模型，尤其是在滚动，俯仰和高度上，相对于恒定速度运动和笛卡尔测量模型，可显着提高进程。
This paper presents an approach to radar odometry on $SE(3)$ which utilizes a constant acceleration motion prior. The motion prior is integrated into a sliding window optimization scheme. We use the Magnus expansion to accurately integrate the motion prior while maintaining real-time performance. In addition, we adopt a polar measurement model to better represent radar detection uncertainties. Our estimator is evaluated using a large real-world dataset from a prototype high-resolution radar sensor. The new motion prior and measurement model signifcantly improve odometry performance relative to the constant velocity motion prior and Cartesian measurement model from our previous work, particularly in roll, pitch and height.
