Positioning Based on Tightly Coupled Multiple Sensors: A Practical Implementation and Experimental Assessment

Abstract

During the last decade, the number of applications for land transportation that depend on systems for accurate positioning has significantly increased. Unfortunately, systems based on low-cost global navigation satellite system (GNSS) components harshly suffer signal impairments due to the environment surrounding the antenna, but new designs based on deeper data fusion and on the combination of different signal processing techniques can overcome limitations without the introduction of expensive components. Supported by a complete mathematical model, this paper presents the design of a real-time positioning system that is based on the tight integration of extremely low-cost sensors and a consumer-grade global positioning system receiver. The design has been validated experimentally through a series of tests carried out in real scenarios. The performance of the new system is compared against a standalone GNSS receiver and surveygrade professional equipment. The results show that a carefully designed and constrained integration of low-cost sensors can have performance comparable to that of an expensive professional equipment.

Keywords

• Global positioning system (GPS)
• inertial navigation system (INS)
• position accuracy
• tight architecture