Inside GNSS Media & Research

JUL-AUG 2019

Issue link: https://insidegnss.epubxp.com/i/1148308

Contents of this Issue

Navigation

Page 48 of 67

www.insidegnss.com J U L Y / A U G U S T 2 0 1 9 Inside GNSS 49 onstrates the use of a LiDAR scanner to describe the perceived environment. According to the relative positions between the detected obstacles and ego-vehicle, their boundaries can be represented in the GNSS skyplot. us, we propose a new GNSS/LiDAR inte- gration scheme to aid the GNSS single point positioning (SPP) as shown in Figure 3. Both the 3-D LiDAR and INS are employed to help the SPP. How can this perceived environment be used to aid GNSS SPP? e intuitive idea is to exclude the NLOS affected measurements from the GNSS position- ing. e exclusion of NLOS measure- ment is very eff ective in less-urbanized areas. It is not the case in deep urban canyons. e dilution of precision (DOP) w ill be easily dis- torted and enlarged if excluding all the NLOS measurements received in the urban c a nyons a s i l lu s- trated in Figure 4. Obviously, enormous NLOS measurements are received and only five LOS measure- ments. If we exclude all the NLOS measurements, the GPS SPP will deteriorate from 92 to 169 meters and HDOP is increased from 0.9 to 3.15. This discussion has been well-discussed in 3-D mapping aided (3DMA) GNSS (Groves 2016, Additional Resources). In fact, the famous GNSS shadow matching is proposed to deal with the same problem. The shadow FIGURE 1 The sensor kit used in the research vehicle of Hong Kong Polytechnic University (PolyU). FIGURE 2 Illustration of real-time perceived environmental features in the perspective of GNSS positioning. Circle plot denotes the Skyplot, which is the top view from the zenith to the vehicle. FIGURE 3 Conventional sensor integration and the proposed integration for autonomous vehicular localization. FIGURE 4 Example of the issue of GNSS NLOS exclusion in an urban canyon at Hong Kong. (a) Environment. (b) Satellites in the skyplot with building boundaries. (c) GNSS positioning results before and after NLOS exclusion. a b c

Articles in this issue

Links on this page

view archives of Inside GNSS Media & Research - JUL-AUG 2019