A direct time-of-flight image sensor with in-pixel surface detection and dynamic vision
This addresses data bottlenecks for outdoor LIDAR applications in self-driving cars, robotics, and AR, representing an incremental improvement in sensor design.
The paper tackled the problem of high data processing demands in 3D flash LIDAR sensors by developing a 64x32 pixel direct time-of-flight imager with in-pixel histogramming, achieving frame rates up to 10 kFPS or 100 kFPS for depth readings and reducing power and processing needs.
3D flash LIDAR is an alternative to the traditional scanning LIDAR systems, promising precise depth imaging in a compact form factor, and free of moving parts, for applications such as self-driving cars, robotics and augmented reality (AR). Typically implemented using single-photon, direct time-of-flight (dToF) receivers in image sensor format, the operation of the devices can be hindered by the large number of photon events needing to be processed and compressed in outdoor scenarios, limiting frame rates and scalability to larger arrays. We here present a 64x32 pixel (256x128 SPAD) dToF imager that overcomes these limitations by using pixels with embedded histogramming, which lock onto and track the return signal. This reduces the size of output data frames considerably, enabling maximum frame rates in the 10 kFPS range or 100 kFPS for direct depth readings. The sensor offers selective readout of pixels detecting surfaces, or those sensing motion, leading to reduced power consumption and off-chip processing requirements. We demonstrate the application of the sensor in mid-range LIDAR.