Less Is More–The Potential of Partial IMUs for Land Vehicle Navigation
A general decline in the cost, size, and power requirements of electronics is accelerating the adoption of integrated GNSS and inertial technologies in consumer applications such as automotive and personal navigation systems. The combination also helps solve the problem of short-term GPS signal outages. Nonetheless, market-driven pressure to lower the cost of materials even further has researchers looking for ways to eliminate additional components from product designs. One possibility is to drop one or more of the relatively expensive gyroscopes from microelectromechanical system (MEMS) versions of inertial measurement units (IMUs). The adoption of other positioning data from the vehicle itself may just help make that possible.
GPS provides reliable long-term navigation information but requires a direct line of sight between the GPS receiver and GPS satellites. On the other hand, an inertial measurement unit (IMU) offers continuous autonomous navigation information, but its accuracy degrades over time due to the cumulative errors of the inertial sensors.
The integration of GPS’s long-term stable accuracy with the continuous but short-term accuracy of an inertial navigation system (INS) can provide accurate and uninterrupted positioning for many difficult navigation scenarios.
Land vehicle navigation is, in fact, one of the most challenging navigation modes because GPS signal quality degrades significantly when a vehicle goes through urban canyons or forest canopies.
Moreover, GPS signals — and, consequently, positioning information — can be completely lost when, for example, a vehicle goes through a tunnel or passes under a bridge.
With the development of low-cost, microelectromechanical (MEMS) inertial sensors and GPS technology, integrated INS/GPS systems are beginning to meet the growing demands of lower cost, smaller size, and seamless navigation solutions for land vehicles.
Although MEMS inertial sensors are very inexpensive compared to conventional sensors, their cost (especially MEMS gyros) is still not acceptable for many low-end civilian applications, such as commercial car navigation or personal location systems.
In an ideal case, three accelerometers and three gyroscopes — that is, a complete IMU — should be part of the IMU to faithfully portray the three-dimensional motions of a vehicle. However, the unremitting drive to reduce the cost to consumers of vehicle navigation systems has led to the development of simplified sensor configurations.
Gyroscopes are usually more expensive than accelerometers in an IMU MEMS. For example, when purchased in bulk, a MEMS gyro chip currently costs about $10 per axis or a total of $30 for a three-dimensional IMU, not including other hardware, electronics, or digital signal processor, which is not acceptable for many applications.
An efficient way to reduce the expense of these systems is to reduce the number of gyros and, therefore, to use a partial IMU (ParIMU) configuration.
For land vehicular use, the most important gyroscope is the vertical gyro that senses the heading of the vehicle. As a result, considerable research has recently been directed towards finding ways to minimize gyroscope usage in the INS or even developing gyro-free INS systems.
The most commonly used ParIMU consists of one vertical gyroscope to detect the heading and two or three horizontal accelerometers for determining the velocity of a vehicle. The accuracy of such partial IMUs will definitely be less than the accuracy of a full IMU.
Consequently, the following key questions arise: is the degradation in accuracy acceptable considering the reduction of cost, can we use the same full IMU processing techniques with ParIMUs, and can we improve the accuracy of the partial IMU using other aiding sources during GPS signals outages?
This article will address these questions based on some of the recent development and testing conducted by the Mobile Multi-Sensor Systems (MMSS) Research Group at the University of Calgary.
In particular, we will compare the use of one- and two-gyroscope IMUs integrated with GPS as well as the potential benefit of exploiting other data on vehicle dynamics, such as the odometer outputs, to improve positioning solutions while minimizing costs.
Summary and Conclusions
Testing of a ParIMU configuration consisting of one vertical gyro and two horizontal accelerometers (1G2A) can reduce the inertial sensor cost by 50 to 65 percent, while maintaining an acceptable navigation performance with some external aiding (NHC and odometer update) which proved to be a very promising design for land-vehicle navigation applications.
(For the complete article, including figures, charts, and images, please download the PDF version at the link above.)
The author would like to thank (1) his former Ph.D. student, Dr. Eun Shin as
ManufacturersThe toolbox software used to analyze the results of the research described in this article is the Aided Inertial Navigation System Toolbox (AINS) for MatLab Software developed by the Mobile Multi-Sensor Systems (MMSS) Research Group at the University of Calgary. An OEM 4 receiver from NovAtel, Inc., Calgary, Alberta, Canada, coupled with the CIMU navigation-grade IMU from Honeywell Inc. was used as the benchmark “truth” system for the tests. The full and partial IMUs developed by the MMSS group incorporated ADXL105 accelerometers and ADXRS150 gyroscopes from Analog Devices Inc. (ADI), Norwood, Massachusetts, USA.
Copyright © 2017 Gibbons Media & Research LLC, all rights reserved.