An Integrated GNSS/INS/DR Positioning Strategy Considering Nonholonomic Constraints for Intelligent Vehicle

The realization of decision-making, trajectory planning, and motion control of intelligent vehicles highly depends on the precise and continuous position information. To provide reliable pose information during GNSS (Global Navigation Satellite Systems)outage, an integrated position strategy based on the GNSS, INS (Inertial Navigation System), DR (dead reckoning) considering nonholonomic constraints (NHC) was proposed in this paper. First, a positioning algorithm of GNSS/INS was established using an Error State Kalman Filer (ESKF). Nonholonomic constraints and vehicle speed were used to correct the navigation error to improve positioning accuracy. Then, the available time of the algorithm during GNSS signal outage was further extended by designing a DR system based on vehicle dynamics. An Extended Kalman Filter (EKF) was used to process the nonlinear system and the information of wheel speed and front wheel angle were integrated into the system. Finally, the fusion positioning algorithm was validated by NI HIL platform. The results showed that, during GNSS outage, the positioning strategy proposed in this paper could reduce the root means square errors (RMSE) of positioning accuracy, compared to the GNSS/INS positioning algorithm, by 78.2% along x-axis and 83.0% along y-axis.