INS Analysis and Navigation in GPS Restricted Environments



The GPS and Vehicle Dynamics Lab at Auburn University is currently working in development and analysis of navigation systems that provide robust solutions to vehicle state when GPS information is fully or partially unavailable. Part of this research focuses on the purely dead-reckoning solution with an Inertial Measurement Unit (IMU), such as when GPS information is fully unavailable. As the IMU output must be numerically integrated to ascertain vehicle attitude, position, and velocity, the uncertainty in these states grow with time. Current work involves development of analytical and empirically-derived expressions that describe the characteristic error growth of the integrated IMU in the dead-reckoning navigation scenario. Knowledge of the error growth as a function of known sensor specifications provides a firm quantification of the accuracy vs. time of the vehicle state when GPS is fully unavailable.

Existing approaches in software allow the vehicle to integrate partial GPS information with an IMU into a GPS/INS system. GPS/INS system advantages include the aiding of the dead-reckoning solution with partial GPS information such as when less than four satellites are in view. The system providing this benefit is commonly known as the tightly-coupled GPS/INS integration scheme. Auburnís GPS and Vehicle Dynamics Lab has current focus to quantify this more robust approach in different navigation scenarios. This quantification provides an accurate estimate of vehicle navigation performance using the tightly-coupled integration approach. An array of test-bed vehicles and various grade GPS and INS systems are available within the lab to test the advanced algorithms in GPS-restricted environments both in post-processing and real-time.

Figure 1:


Inertial Measurement Integration Error Growth for KVH-5000 Fiber Optic Gyro

Figure 2:


Attitude Error for Integrated IMU data for simulated consumer-grade inertial measurements