1. Positioning
Positioning in unmanned driving for ten minutes is important. Only with real-time location information can the system make the next interpretation, decide where to go, and how to get there. There are many ways to position it now. Such as carrier phase difference technology (RTK), but RTK is still subject to signal interference. Especially in some cities, buildings and trees, as well as tunnels and tunnels, its signal is easily interrupted. At the same time, sensors such as cameras are also used to sense the external environment, construct an environmental model, and use the model to determine the location of the vehicle, but its dependence on the environment is relatively strong, such as backlighting or rain and snow, such positioning is prone to failure. Lidar relies on comparing the initial position of the vehicle with the high-precision map information to obtain a precise position. First, sensors such as GPS, IMU, and wheel speed give an initial (probably) position. Secondly, the local point cloud information of the lidar is extracted, and the vector features in the global coordinate system are obtained by combining the initial positions. Finally, the vector features of the previous step are matched with the feature information of the high-precision map to obtain accurate global positioning. Therefore, in terms of positioning, the use of laser radar has an unparalleled advantage in terms of accuracy and stability. The only shortcoming is that the current production cost of laser radar is high. On the other hand, technological innovation is carried out in the solid state direction, and it is developing toward the low-cost and mass production of solid state. Many domestic and foreign manufacturers are accelerating. Innovation, cost will not become a very difficult problem in the near future.
2. Obstacle detection and classification
For obstacle detection and classification, there are currently applications of vision and lidar, which are not in conflict. Lidar does not rely on illumination, its viewing angle is 360 degrees, the calculation is relatively small, and it can be scanned in real time. It is generally used within 100 milliseconds. In the process of scanning, the laser radar first identifies obstacles, knows the position of the obstacle in space, and then classifies according to the existing obstacles. For example, cars and people, we divide these obstacles into independent individuals, and then separate and separate the individual to match, so as to classify the obstacles and track the objects. The process of tracking, first of all, is to divide the point cloud, and to make the associated target through the point cloud. We know whether the last one and the next one belong to the same object, then perform target tracking and output target tracking information.
3. For advanced driver assistance systems (ADAS)
The Advanced Driver Assistance System (ADAS) uses a variety of sensors installed in the vehicle to collect environmental data inside and outside the vehicle at the first time to identify and detect static and dynamic objects. Technical processing such as tracking, which allows the driver to detect the dangers that may occur in the fastest time, to attract attention and improve the safety of active safety technology. If the lidar can effectively control the cost, the driver assistance function with a lower ADAS rating also requires a lidar. The reason is that camera-based ADAS and driverless systems, or the use of millimeter waves alone, are highly limited. The first is the problem of the angle of view. In order to ensure a sufficient detection distance, the angle of the field of view should not be too large, and this leads to a very large lateral blind zone. In response to this phenomenon, some companies have designed multiple cameras to solve such problems, but the results are not satisfactory. The same multi-camera camera will have overlapping areas and will increase the difficulty of processing data. The second is the low-speed problem. It turns out that at low speeds, the performance of the camera is not satisfactory, and even the target of slow or static movement is difficult to identify. So laser radar has great potential in ADAS.
Mobile：
Telephone：
