Low-Cost Computational Sensors that Replace Lidar Sensors

About this Session
Existing scanning lidar systems are a cornerstone imaging modality in fully self-driving vehicles systems. While directly measuring depth via round-trip pulse travel, lidar systems are limited to low spatial resolution at large ranges due to mechanically limited angular sampling rates, restricting scene understanding tasks to close-range clusters with dense sampling. Moreover, today’s pulsed lidar scanners suffer from high cost, power consumption, large form-factors, and they fail in the presence of strong backscatter.

We depart from point scanning and demonstrate that it is possible to turn a low-cost CMOS gated imager into a dense depth camera with a 150 m range. The method is real-time, handles back-scatter, and provides dense depth at long ranges. We will share validation results from over 10,000 km field testing in northern Europe.

We will also demonstrate how to extract multi-bounce information from low-cost Doppler Radar measurements, allowing vehicles to see around corners. In contrast to lidar-based methods, we increase ranges from meter-sized scenes to 50m distances. Our approach is the first practical non-line-of-sight detection and tracking method.

Algolux Portal


Contact Speaker