Infrared Reflectance Requirements of the Surrogate Grass from Various Viewing Angles 2019-01-1019
Road Departure Mitigation (RDM) is a promising active safety features being considered for vehicles. As the grass road edge is the most commonly seen road edge, RDM systems should be able to detect grass road edges by using onboard sensors when lane markings are poor or absent. Standard testing of RDM on a road with real grass as road edge has consistency issue, since the grass color may be different at different test sites and in different seasons. A solution is to develop surrogate grass that has the same characteristics of the representative real grass that provides consistent test environment anywhere and anytime from the viewpoint of commonly used sensors in RDM systems. LIDAR is one such popular sensor that uses near infrared (IR) laser beams to detect the range and reflectance of objects. In this paper, we describe the infrared reflectance behavior of real grass and suggest the IR requirement of surrogate grass for different viewing angles of LIDAR.
Since most automotive LIDARs use laser beams that are typically in the spectral range between 800 nm and 1600 nm, the spectral reflectance of grass within this spectral region was studied. To do so, the reflectance spectra of grass were measured at different measurement angles between 0° (normal to the object surface) to 70° with a spectrometer with the 350-2500 nm wavelength range. The illumination angle was kept equal to the measurement angle during measurement to simulate LIDAR operation conditions. It was found that reflectance is lowest at 0° measurement angle and highest at 70° measurement angle. In general, grass reflectance increases as the measurement angle increases. The height of grass does not affect the IR reflectance much. Based on the measurement results, the suggested representative reflectance of surrogate grass for RDM testing is generated. The suggested IR reflectance for the surrogate grass represents normal daytime conditions. The IR reflectance of wet grass in the early morning should be much lower.
This systematic IR requirement study is the part of the first step towards making consistent surrogate grass, which is essential for safely testing the RDM for all levels of driving automation.
DAN SHEN, Lin Li, Abir Saha, Stanley Chien, Qiang Yi, Rini Sherony
Indiana Univ Purdue Univ Indianapolis, Toyota Motor North America