Navigation Errors Introduced By Ground Vehicle Dynamics

An analysis of navigational accuracy when influenced by ground vehicle dynamics is presented. Tests beds outfitted with various sensor suites were used to collect data when normal and extreme driving maneuvers are executed. The data was run through an extended Kalman filter to produce a navigation solution. The Kalman filter inputs varied on each test bed, using both automotive and tactical grade Inertial Measurement Units (IMU). The position, velocity, and course measurements were obtained from a DGPS unit mounted on the vehicles and used as a truth measurement when exploring dead reckoning error. Additional measurements, such as wheel speed, radar speed, and magnetometer heading, were added to improve the robustness and reliability of the solution. The results of the work show the effect of both longitudinal and lateral vehicle slip on the navigation solution. In addition, the attempt of the various sensors to correct the errors is investigated.