학술논문

Personal Inertial Navigation System Assisted by MEMS Ground Reaction Sensor Array and Interface ASIC for GPS-Denied Environment
Document Type
Periodical
Source
IEEE Journal of Solid-State Circuits IEEE J. Solid-State Circuits Solid-State Circuits, IEEE Journal of. 53(11):3039-3049 Nov, 2018
Subject
Components, Circuits, Devices and Systems
Engineered Materials, Dielectrics and Plasmas
Computing and Processing
Navigation
Electrodes
Sensor arrays
Pins
Capacitance
Timing
C<%2Fitalic>%2FV<%2Fitalic>%29+converter%22">Capacitance-to-voltage (C/V) converter
capacitive sensor
cyclic analog-to-digital converter
GPS-denied navigation
ground reaction sensor array (GRSA)
inertial measurement unit (IMU)
personal inertial navigation system (PINS)
pressure sensor array
switch capacitance compensation
tactile sensor array
Language
ISSN
0018-9200
1558-173X
Abstract
A personal inertial navigation system (PINS) assisted by a microelectromechanical systems (MEMS)-based $13 \times 26$ ground reaction sensor array (GRSA) and a low-power interface application-specified integrated circuit (ASIC) has been designed and demonstrated for GPS-denied environment. The GRSA operating in a contact mode achieves a sensitivity of approximately 3.7 fF/kPa at each sensor node. An electronic interface system, consisting of a capacitance-to-voltage ( ${C}/{V}$ ) converter followed by a correlated double sampling stage, is designed to convert the GRSA capacitance change to an analog output voltage. The analog output voltage is then digitized by a 12-bit cyclic analog-to-digital converter (ADC). Switch capacitance compensation technique is employed to ensure the ADC performance. The ASIC is fabricated in 0.35- $\mu \text{m}$ CMOS process and dissipates a power of 3 mW. The prototype system incorporates a GRSA, an ASIC, and a commercial nine degree-of-freedom (DOF) inertial measurement unit (IMU) in the heel region of a boot. The GRSA can determine an accurate foot-on-ground timing based on the pressure profiles detected during walking, thus enabling an accurate position calculation and a precise zero velocity update. Furthermore, a system calibration procedure measures the IMU inherent directional drift and scaling factor errors, and compensates them for the navigation data to achieve a superior performance. The prototype system demonstrates a position accuracy of approximately 5.5 m over a navigation distance of 3100 m. The prototype system also achieves a consistent performance over different field tests with various distances and random paths. System characterization results further indicate a tradeoff between sensor array size and system resolution for a given navigation performance requirement, thus providing a design guideline for future system optimization.