Abstract: Autonomous driving has become an important direction for the development of future automotive technology. However, the lack of perception accuracy of selfdriving cars at this stage has become an important factor restricting the application of selfdriving cars. In order to solve the above problems, this paper proposes an adaptive data association method based on the theory of multisensor information fusion, which considers the influence of the sensor's error characteristic model and the target's motion state on the data association to achieve target tracking in a cluttered environment. Design experiments to verify the effectiveness of the method. The experimental results show that the fusion sensing results proposed in this paper can effectively reduce the error value, and the target trajectory tracking method can 100% guarantee the consistency of the target number in all experimental scenarios.

Key words: Multi-sensor fusion; Target tracking; Adaptive tracking gate

摘要： 自动驾驶已经成为未来汽车技术发展的一个重要方向。但现阶段自动驾驶汽车的感知精度 不足已经成为限制自动驾驶汽车应用的一个重要因素。为解决上述问题，文章基于多传感器信息 融合理论，提出一种自适应数据关联方法，分别考虑传感器的误差特性模型、目标的运动状态对 数据关联的影响实现杂波环境中的目标追踪。并实验验证方法的有效性，实验结果表明，文章提 出的融合感知结果能够有效地降低误差值，且目标轨迹追踪方法在所有实验场景中能 100%保证目 标编号的一致性。

关键词: 多传感器融合；目标追踪；自适应跟踪门限