Design and verification of parking space detection based on AMR sensor

Parking space detection is crucial to vehicle parking guidance and traffic management in large parking areas. The detection from geomagnetic sensor detector is among the lowest cost technologies commonly used for collecting such data. Like many parking space detecting technologies, the geomagnetic sensor detector approach works well under normal circumstances but is subject to errors under specific situations. To address this problem, we propose a new approach to detect vehicle states in whole parking process, which is divided into 4 classes, to make parking space detection more accurate and flexible. From this approach, the concept of the detector based on three component anisotropic magneto resistive (AMR) sensor is proposed, a comprehensive illustration on how the detector works is presented, and an algorithm to recognize 4 kinds of vehicle states is discussed. Pilot testing experiments proves that the approach can provide efficient parking space detection under certain interference.