Single-Photon Detectors Modeling and Selection Criteria for High-Background LiDAR

摘要

Vision systems capable of acquiring both two-dimensional and three-dimensional information through Light Detection And Ranging are assuming ever-increasing importance, being this market driven by the push from automotive companies to develop systems to be integrated in self-driving vehicles. Among others, candidate sensors for these systems are avalanche photodiodes, single-photon avalanche diodes, and silicon photomultipliers. Avalanche Photodiodes provide a good robustness to high background light at the cost of requiring an analog readout, instead Single-Photon Avalanche Diodes offer the possibility to implement digital readout and single-photon sensitivity, but are prone to saturation at extremely high background levels. We compare these three single- and multi-photon detector topologies, operated either in linear or digital regime, aiming at identifying the best suited detector to achieve the highest performance in Light Detection And Ranging applications at the lowest optical power active illumination and in presence of intense background (e.g. 100 klux). We present Matlab modelling and simulations and their experimental validation. Eventually, we propose a nomogram (referred to 100 m target distance) for identifying the most suited sensor topology across different operating areas and constraints, in order to achieve at least 70 success ratio.