GPS is indispensable to the current vehicle positioning technology. Due to the error of GPS, multipath and update frequency is low, we can't just rely on GPS positioning. Inertial sensor has high update frequency, it can be complementary with GPS. Using sensor fusion technology, we can be the integration of GPS and inertial sensor data, from each director, in order to achieve better location performance.
Inertial navigation system (INS) detecting acceleration and rotation motion of high frequency (1 KHZ) sensor, the inertial sensor data processing after the displacement and rotation of the vehicle can drawn real-time information. INS have bias and noise problems affect the outcome. By using a sensor fusion technique based on kalman filtering, we can be the integration of GPS and inertial sensor data, from each director, in order to achieve better location performance. Attention because of the unmanned for reliability and security requirement is very high, so based on GPS and inertial sensor positioning is not the only way of positioning in the unmanned, we also use LiDAR point cloud and high precision map matching, and locating methods such as visual mileage calculation method for various positioning method to correct each other to achieve more accurate results.