In this era of mobile Internet, whether it is location sharing, surrounding search or route planning, it is necessary to locate users and obtain the location of users, and these operations are inseparable from location services. However, when everyone is sharing a location, it often happens that the path is offset and the location positioning is inaccurate due to building occlusion. The SuperGNSS technology of Huawei Location Services is solving this problem for everyone and is committed to providing you with more convenient and accurate location services.
Before introducing the principle of SuperGNSS technology, let me explain a term-urban canyon. It can be seen from this picture that as the tall buildings in the city become higher and denser, when we look up, we can only see a piece of sky or a line of sky. This kind of street cuts the surrounding dense buildings. Man-made canyons formed by blocks are called urban canyons. Well-known examples such as the Kowloon and Central areas of Hong Kong, China, and Manhattan, New York, USA, have such phenomena in more than 100 cities around the world.
So, why is satellite positioning inaccurate in the urban canyon scene? The reason is that when the satellite signal propagates from the sky to the mobile phone, reflection of the signal path will occur due to the building block. In this picture, the blue path is reflected into the mobile phone once, and the yellow path is more serious because it is reflected back to the mobile phone. inner.
In fact, in the dense street signal scene of an actual city, most of the signals pass through a non-straight path and are reflected multiple times before reaching the mobile phone, which causes our satellite positioning to have an offset of 50 to 200 meters.
For example, when using a taxi-hailing application, when people stand at an intersection and initiate a taxi, the positioning often cannot accurately obtain everyone's location. Originally, you were at the northeast corner of the intersection, but the positioning was set at the southwest corner. At this time, the driver could not Find your real location; and when using life apps, when people search for nearby locations, there is a gourmet shop in your location, but because of the path offset and positioning results, it is displayed in the next block. Then when you go to the next block, you will find that you can't find this gourmet shop. The above-mentioned positioning deviations often greatly reduce the user experience of using the application.
So how does SuperGNSS technology solve such problems? First, use the ephemeris signal to calculate the satellite's current three-dimensional coordinates in the sky. On the other hand, the GNSS original observation scalar calculates the actual position of the satellite. With these data, we can calculate the propagation model of the satellite signal to the mobile phone. Then establish a ray tracing algorithm to track how the satellite signals in the sky enter the mobile phone, and use the three-dimensional model library on the ground to predict the trajectory of the satellite signal, and use calculations to turn these trajectories into a direct pseudorange. To improve positioning accuracy.
To what extent? SuperGNSS technology can completely offset the effect of building occlusion. First of all, we have a model of deep learning for target recognition. Based on this model, we can fully learn the bottom contour of a building, including the coordinates of the building. Then use the multi-spectrum analysis technique in remote sensing measurement to calculate the height of the building. With the outline plus the height, our 3D model library is constructed. Based on this 3D model library, the final pseudorange can be calibrated. Now our 3D model library can achieve meter-level accuracy.
There are many scenarios where SuperGNSS technology is applied. In addition to tracking the bilateral occlusion of the primary reflection, the optimization accuracy of the secondary reflection tracking algorithm has also been greatly improved, such as single-sided occlusion, crossroads, and T-junctions. Can also be accurately identified. The accuracy of the primary reflection tracking can reach 90%, and the accuracy of the secondary reflection can reach 75%.
Let's take a look at the comparison tests we have done with competing products in Tokyo and Frankfurt. This mainly tests the accuracy of the positioning on the left and right sides of the road. In this scenario, the accuracy of Mate30's road recognition on both sides of the road using SuperGNSS technology is nearly 4 times that of competing products, up to 90.77%.
In addition, we have conducted field tests around the world, which are the test results of some typical cities. There are Abu Dhabi in the UAE, Barcelona and Rome in Europe, Santiago in Chile in South America, and Beijing and Shanghai in China. The blue point in this picture represents the user's real location, the green point is the positioning result of our SuperGNSS, and the red is the positioning result without our technology. Judging from these figures, our positioning is very accurate.
Finally, if you want to access Huawei's location service, you only need five simple steps to complete. If you want to call SuperGNSS, you don't need to do any extra work at all, just call our location service interface to support it, including database preparation, positioning calculations, etc., all hidden behind this interface.
For more details, please refer to:
Huawei Developer Alliance official website:https://developer.huawei.com/consumer/cn/hms?ha_source=hms1
Get the development guide document:https://developer.huawei.com/consumer/cn/doc/development?ha_source=hms1 To
participate in the developer discussion, please go to the Reddit community:https://www.reddit.com/r/HuaweiDevelopers/ To
download the demo and sample code, please go to Github:https://github.com/HMS-Core to
solve integration problems, please go to Stack Overflow:https://stackoverflow.com/questions/tagged/huawei-mobile-services?tab=Newest
Original link:https://developer.huawei.com/consumer/cn/forum/topic/0201441829931230425?fid=18
Author: Pepper