Recent Developments in Laser-Driven and Hollow-Core Fiber Optic Gyroscopes

Abstract

The interferometric fiber optic gyroscope (FOG) has been a successful commercial product for over 25 years, and it is used in a variety of applications, including ship and sub-sea inertial navigation, along with stabilization and positioning. Despite these significant achievements, improvements to the FOG in terms of performance and cost are required to broaden its applicability to other markets, including navigation of aircraft. One area of weakness is the scale-factor stability of conventional FOGs, which is typically limited to 10–100 ppm, compared to the 1–5-ppm stability required for the navigation of an aircraft or a submarine. The scale factor of a FOG is proportional to the reciprocal of the wavelength of the Er-doped superfluorescent fiber source (SFS) used to interrogate it. An SFS emits light in a broad bandwidth of typically a few tens of nanometers. Historically, it has proven difficult to stabilize the mean wavelength of this radiation to better than 10 ppm, leading to relatively poor scale-factor stability. In addition, the noise of an SFS-driven FOG is limited by the relatively large excess noise, or relative intensity noise (RIN), of the SFS, which originates from amplified spontaneous emission. Techniques have been developed both to stabilize the mean wavelength of an SFS to the sub-ppm level and to subtract the excess noise. However, these solutions add significant complexity and cost, making the FOG less competitive with other technologies, such as the ring laser gyro (RLG). The two primary limitations of commercial FOGs therefore originate solely from the use of an SFS. A somewhat less significant issue is thermal transients in the FOG sensing coil, which induce a phase error known as the Shupe effect that can limit the bias stability of the FOG. This error has been substantially overcome through quadrupolar coil winding, a special winding procedure designed to ensure that sections of fiber that are symmetrically located with respect to the mid-point of the sensing loop are physically in close proximity on the fiber spool. Despite its successes, this solution is not always reproducible, and it does not reduce the Shupe error sufficiently for high-end applications. Finally, another minor problem is the optical nonreciprocity introduced by exposure of the fiber coil to the earth’s varying magnetic field, which produces a phase bias error in the fiber coil via the Faraday effect. This issue can be effectively mitigated by shielding the coil in a mu-metal enclosure, though this solution increases the weight, size, and cost of the sensor.