The GPS Vehicle Dynamics Laboratory focuses on the control and navigation of vehicles using GPS in conjunction with other sensors, such as Inertial Navigation System (INS) sensors. The laboratory has several research thrusts including: sensor fusion/integration, on-line system identification, adaptive and robust control algorithms, and vehicle state and parameter estimation. These research thrusts are focused towards vehicle dynamics and transportation, including heavy trucks, passenger cars, off-road vehicles, as well as autonomous and unmanned vehicles. The laboratory consists of various GPS receivers (including a software GPS receiver), Inertial Measurement Units (IMUs), an instrumented Chevrolet Blazer, an automatically steered John Deere tractor, and access to an iRobot ATRV. Current projects include ultra-tight GPS/INS coupling (sponsored by the Army), study of vehicle rollover propensity, improved steering control of GPS guided farm tractors (sponsored by John Deere), vehicle and driver monitoring, and navigation and control of unmanned ground vehicles (UGVs).

As part of a current project with U.S. Army Aviation and Missile Command, a GPS receiver capable of performing ultra-tight GPS/INS integration has been acquired. Ultra-tight GPS/INS coupling provides improved anti-jamming resistance, improved satellite tracking, as well as allows for immediate GPS signal reacquisition after short GPS outages. This GPS unit could be valuable in investigating GPS navigation in cluttered environments, where GPS satellite signals become unavailable and available for intermittent periods. The GPS Vehicle Dynamics Laboratory also has access to the National Center for Asphalt Testing (NCAT) test track (http://www.pavetrack.com/). Validation of navigation and parameter and state estimation algorithms can be performed using the vehicles on the NCAT track. GPS and inertial sensors can be mounted on the semi-trucks or our own test vehicles to validate proposed estimation and control algorithms. Additionally, errors such as jamming, multi-path, and other sensor errors can be simulated to test the algorithms ability to reject these disturbances and continue to provide an accurate navigation solution.