Adaptive scanning for Lidar sensors based on three rotating prisms

Lidar systems are essential sensors used in high level autonomous driving. In scanning Lidar systems, collimated beams are steered by optical elements to cover a desired field of view (FOV). The distance between two adjacent scanning points refers to the ability that the system is able to detect small objects. By changing the size and orientation of the FOV and the scanning point density an adaptation for different driving situations is possible. In this way the detection of pedestrians e.g. in an urban scenario or of small obstacles in great distances on motorways can be improved. In this paper, an adaptive scanning method based on three rotating prisms is simulated and discussed. After figuring out the underlying optical principle, scan patterns at different rotational speed ratios and phase differences of the prisms are simulated. Based on the results we discuss the applicability of the presented concept for different driving scenarios.