Speaker
Description
LiDAR sensors with fast sample rates and precise spatial resolution are an important enabling technology for autonomous driving and unmanned aerial vehicles. With ubiquitous use of the technology, robustness to temporal and spectral congestion is becoming a key criterion for LiDAR systems. FMCW LiDAR using chaotic micro-combs has previously been shown to be a promising approach for accurate high-speed 3D sensing in the presence of interfering signal.
Here, we investigate dispersively parallelised PRBS LiDAR as an alternative approach for micro-comb-enabled LiDAR technology in congested environments. Our method, which employs amplitude-encoded PRBS pulses, performs signal modulation and temporal cross-correlation for all optical comb lines with two electro optical amplitude modulators. By performing the cross-correlation in the optical domain through FPGA-controlled shifts of the encoding and decoding PRBS sequences, the need for ultrafast detectors and data acquisition / processing systems is removed.
We demonstrate the principle in a laboratory scale setup using a microcomb operating with a repetition rate of 100GHz at 1555nm wavelength. Combining the parallel dispersive sampling line with an angular mirror rotation across the orthogonal axis we perform 2D depth scans of a 3D target and characterize linearity, resolution and repeatability of our measurements.
The resource-efficient frequency multiplexing using a microcomb and cross-correlation in the optical domain, could allow for massive parallelization of our digital LiDAR system. Using code division multiplexing for different sensors in LiDAR congested environments, provides an additional degree of freedom for the presented approach to increase the robustness of signals against detrimental LiDAR interferences.
| I am the presenting author | Yes |
|---|