多模式切换下车载雷达数据融合目标跟踪算法

Vehicle radar data fusion target tracking algorithm with multi-mode switching

为获得更全面的感知信息,自动驾驶系统常采用FMCW毫米波雷达等多种传感器对道路环境进行感知。汽车目标跟踪是车载FMCW毫米波雷达的主要研究方向,该雷达常采用远近多模式切换方案以兼顾测量精度和测量范围。为获得相较于单模式雷达更高的估计精度,需要对多模式FMCW雷达中的测量数据进行数据融合,并应用于目标跟踪算法中。提出一种多模式切换下的车载FMCW毫米波雷达数据融合目标跟踪算法,在仿真车载场景和道路实测数据进行了验证。实验结果表明该算法相较于单模式毫米波雷达目标跟踪算法,具有更高和更稳定的跟踪精度。

In order to obtain more comprehensive sensing information,automotive driving systems often use a variety of sensors such as FMCW millimeter wave radar to perceive the road environment.Vehicle target tracking is the main topic research on in-vehicle FMCW millimeter wave radar,which often uses a multi-mode switching scheme between near and far to balance measurement accuracy and measurement range.In order to obtain higher estimation accuracy compared with single-mode radar,data fusion of measurement data in multi-mode FMCW radar is required and applied to the target tracking algorithm.This paper proposes an in-vehicle FMCW millimeter wave radar data fusion target tracking algorithm for multi-mode switching and the algorithm is validated in both simulated and real road measurement tests.The experimental results show that the algorithm has higher and more stable tracking accuracy compared to the single-mode millimeter wave radar target tracking algorithm.