Visual Inertial Odometry (VIO) is a navigation system using Camera and Inertial Measurement Unit (IMU). Nowadays, a camera includes IMU and sensor cost is low. Also, The camera is easily removable. Because of its characteristic, VIO can be easily used even if the moving object is changed. In a feature-rich environment, VIO positioning performance is accurate. Due to these advantages, research on VIO is being actively conducted. VIO has the disadvantage that the positioning performance deteriorates when there is no feature or dark environment. On the other hand, GPS can always obtain measurements day or night in an external environment. Especially, the relative position can be obtained by using Doppler. This is a good measurement because it gives exact position. These good measurements can complement the disadvantages of VIO. We introduce GPS/VIO system based on nonlinear optimization technique. This technique is a method of obtaining the solution of the optimal state variable by entering the relative position measured by VIO, the relative position by Doppler, and the global position by GPS in the nonlinear optimization method.

Keywords: GPS, VIO, nonlinear optimization method