Vehicle road departure mitigation system (RDMS), as new active safety technology, has been introduced into the market in recent years. This system can detect roadside objects and road edges to reduce the risk of roadway departure crashes. To evaluate and improve the performance of RDMS, surrogates of roadside objects, which have the same camera, radar, and LiDAR characteristics of the real objects, need to be developed. Grass is the most common road edge in the U.S. as seen from the real road data. This paper describes the development of surrogate grass. The LiDAR (infrared) and radar characteristics of the selected artificial turf (grass) are obtained and compared with those of real grass. In order to make the surrogate grass match the real grass in the view of sensors (LiDAR, radar and camera), a special color coating with high reflectance material is applied to the artificial turf. Both LiDAR and radar measurements confirmed that the surrogate grass closely match the key characteristics of the real grass. Five grass colors and eighteen color patterns were identified based on 1,021 grass road-edge samples from all states of the U.S. 300-meter long surrogate grass was made and successfully used on the test track for the vehicle RDMS evaluation.
I. INTRODUCTIONRoad departure is a major cause of fatal vehicle crashes in the U.S. [1][2][3][4]. Based on the data from the Federal Highway Administration (FHWA), an average of 19,223 fatalities occurred per year due to the road departure from 2015 to 2017, accounting for 52% of all the traffic fatalities in the U.S. [5]. Considering the severity of the road departure crash, advanced active safety technology, such as roadway departure mitigation system (RDMS), has been developed and introduced into the market in recent years for reducing the road departure crashes [6][7][8][9][10][11][12][13][14][15][16][17][18]. The RDMS is designed to have the capability of detecting the road edges/boundaries and recognizing the roadside objects, even when the lane marking is unclear or does not exist.However, to the best knowledge of the authors, the standard testing procedures and equipment for evaluating the performance of RDMS have not been fully developed. Hence, it is imperative to research and develop methods,