The realization of autonomous driving is highly dependent on core technologies such as environment perception, control execution, high-precision positioning, and path planning. Among them, through environmental perception, real-time and accurate identification of the objects surrounding the vehicle's driving path that may have hidden dangers for its safe driving, providing a reliable source of information for vehicle control execution; real-time positioning of the vehicle position through high-precision maps and high-precision positioning technology, It also provides path planning and scheduling for vehicle driving through big data algorithms.
In some subdivisions, autonomous driving can take the lead in landing, such as sanitation trucks driving at low speeds on fixed routes, container trucks in ports, logistics vehicles in closed communities, indoor parking garages, and shuttle buses in parks. However, the large-scale implementation of autonomous driving still has major challenges in vehicle positioning:
GPS/GNSS data delays are high in high-speed driving conditions. High-speed outdoor movement can rely on GPS/GNSS positioning, and some areas rely on RTK for precise positioning. The mobile station performs real-time carrier phase differential processing (less than one second) on the collected satellite data and the received data link to obtain centimeter-level positioning results. After the vehicle reaches 120KM/H, it moves 33 meters per second, and 100MS is 3.3 meters. In addition, GPS data transmission delay is relatively high. It requires DSRC/V-2X and other technical vehicle-road coordination to support the safety requirements of autonomous driving.
Large-scale indoor or semi-indoor venues are severely blocked, and there are blind areas for GPS/GNSS signals. In scenes such as overpasses, culverts, tunnels, boulevards, underground parking lots, buildings, etc., GPS/GNSS signals cannot be received, and some scenes, such as overpasses, need to locate the upper and lower floors. The existing technology cannot meet such accuracy. It is required that the vehicle cannot be located, and other methods of positioning are required, especially the positioning without blind spots, to ensure the smooth and stable automatic driving of the vehicle.
Autonomous driving relies on high-precision map technology, differential GPS technology and inertial navigation to achieve centimeter-level positioning, which can meet outdoor needs. However, GPS/GNSS has problems with high-speed driving signal delay and signal occlusion. At the same time, inertial navigation will have a large cumulative error for a long time, which becomes the bottleneck of high-precision positioning for automatic driving. Because of this, ultra-wideband UWB positioning technology has entered people's field of vision, becoming the first choice for high-precision positioning technology to make up for the inability of GPS and inertial navigation to locate in complex environments and indoors, and to ensure the integration of indoor and outdoor automatic driving.
By using UWB positioning as an observation method and assisting lidar to evaluate the position and attitude during driving, the reliability of the actual operation of automatic driving is realized. In the deployment of the UWB positioning system, the base station adopts offline power supply installation without networking, and directly calculates the relative coordinates (converted to WSG84 coordinates) through the tags and provides them to the ECU, which can be seamlessly integrated with GPS signals and is reliable as an "indoor GPS" for autonomous driving Location data source. The availability
of UWB positioning technology in the field of autonomous driving has been realized through product architecture innovation and fusion algorithms :
- Broadcast transmission, distributed solution;
- The positioning system has no capacity limitation;
- The positioning response time delay is greatly shortened;
- Topological structure is equivalent to GPS/GNSS;
- Realize the switch between UWB positioning and GPS/GNSS positioning without blind spots;
- Vehicle positioning accuracy reaches the centimeter level;
- Can develop location big data cloud platform to realize real-time location data upload to the cloud;
- The positioning base station is fully wireless and synchronized.
To realize the large-scale landing application of UWB positioning technology in the field of autonomous driving, the prerequisite is the construction of infrastructure such as positioning base stations, and the construction cost is considerable. The "new infrastructure" has become a good opportunity for the deployment of a large number of UWB positioning base stations. UWB positioning technology can effectively compensate for the signal blindness of 5G small base stations. At the same time, smart light poles, as a new public infrastructure of smart cities, have begun to quickly land in all parts of the country. Become an excellent carrier for positioning base stations, which is an opportunity provided by UWB positioning technology in the market development of autonomous driving.