VEHICLE LIDAR POLARIZATION

The present disclosure provides 'vehicle lidar polarization'. A computer, including a processor and a memory, the memory including instructions to be executed by the processor to receive an emitted polarized light beam at a lidar receiver that determines a polarization pattern and a distance to an object, wherein the polarization pattern is determined by comparing a linear polarization pattern and a circular polarization pattern and identify the object by processing the polarization pattern and the distance with a deep neural network, wherein the identity of the object can be metallic or non-metallic. The instructions can include further instructions to operate a vehicle based on the identified object. 本公开提供"车辆激光雷达偏振"。一种计算机，其包括处理器和存储器，所述存储器包括可由所述处理器执行以进行以下动作的指令：在激光雷达接收器处接收发射的偏振光束，所述激光雷达接收器确定偏振图案和到对象的距离，其中通过比较线偏振图案和圆偏振图案来确定所述偏振图案；以及通过用深度神经网络处理所述偏振图案和所述距离来识别所述对象，其中所述对象的身份可以是金属的或非金属的。所述指令可以包括另外的指令：基于识别的对象来操作车辆。