GNSS/Electronic Compass/Road Segment Information Fusion for Vehicle-to-Vehicle Collision Avoidance Application

Abstract

The increasing number of vehicles in modern cities brings the problem of increasing crashes. One of the applications or services of Intelligent Transportation Systems (ITS) conceived to improve safety and reduce congestion is collision avoidance. This safety critical application requires sub-meter level vehicle state estimation accuracy with very high integrity, continuity and availability, to detect an impending collision and issue a warning or intervene in the case that the warning is not heeded. Because of the challenging city environment, to date there is no approved method capable of delivering this high level of performance in vehicle state estimation. In particular, the current Global Navigation Satellite System (GNSS) based collision avoidance systems have the major limitation that the real-time accuracy of dynamic state estimation deteriorates during abrupt acceleration and deceleration situations, compromising the integrity of collision avoidance. Therefore, to provide the Required Navigation Performance (RNP) for collision avoidance, this paper proposes a novel Particle Filter (PF) based model for the integration or fusion of real-time kinematic (RTK) GNSS position solutions with electronic compass and road segment data used in conjunction with an Autoregressive (AR) motion model. The real-time vehicle state estimates are used together with distance based collision avoidance algorithms to predict potential collisions. The algorithms are tested by simulation and in the field representing a low density urban environment. The results show that the proposed algorithm meets the horizontal positioning accuracy requirement for collision avoidance and is superior to positioning accuracy of GNSS only, traditional Constant Velocity (CV) and Constant Acceleration (CA) based motion models, with a significant improvement in the prediction accuracy of potential collision.

Development and Application of a High-Precision Portable Digital Compass System for Improving Combined Navigation Performance

“…Flashing for use in the next test; since the environment, location, time, etc. of each use are different, it is recommended to test after each power-on [ 21 ].…”

Drive Force and Longitudinal Dynamics Estimation in Heavy-Duty Vehicles

“…Different options are explored to record data during driving tests. The typical approach is using an Inertial Sensor Unit (IMU) with Global Positioning System (GPS) to estimate the position and speed of the vehicle [17,18]. However, a different option is possible: combining these IMU/GPS sensors with the low frequency internal information of the vehicle.…”