Design Article
MicroBlaze hosts mobile multisensor navigation system
Walid Farid Abdelfatah, Queen's University; Jacques Georgy, Trusted Positioning Inc.; Aboelmagd Noureldin, Queen's University/Royal Military College of Canada
3/29/2011 7:36 AM EDT
Researchers used Xilinx’s soft-core processor to develop an integrated navigation solution that works in places where GPS doesn’t.
The global positioning system (GPS) is a satellite-based navigation system that is widely used for different navigation applications. In an open sky, GPS can provide an accurate navigation solution. However, in urban canyons, tunnels and indoor environments that block the satellite signals, GPS fails to provide continuous and reliable coverage.
In the search for a more accurate and low-cost positioning solution in GPS-denied areas, researchers are developing integrated navigation algorithms that utilize measurements from low-cost sensors such as accelerometers, gyroscopes, speedometers, barometers and others, and fuses them with the measurements from the GPS receiver. The fusion is accomplished using either Kalman filter (KF), particle filter or artificial-intelligence techniques.
Once a navigation algorithm is developed, verified and proven to be worthy, the ultimate goal is to put it on a low-cost, real-time embedded system. Such a system must acquire and synchronize the measurements from the different sensors, then apply the navigation algorithm, which will integrate these aligned measurements, yielding a real-time solution at a defined rate.
The transition from algorithm research to realization is a crucial step in assessing the practicality and effectiveness of a navigation algorithm and, consequently, the developed embedded system, either for a proof of concept or to be accepted as a consumer product. In the transition process, there is no unique methodology that designers can follow to create an embedded system. Depending on the platform of choice—such as microcontrollers, digital signal processors and field-programmable gate arrays—system designers use different methodologies to develop the final product.
---------------------------------------
Editor's Note: I am delighted to have the opportunity to present the following piece from the Issue 74 (First quarter 2011) of Xcell Journal, and is reproduced here with the kind permission of Xilinx Inc.---------------------------------------
The global positioning system (GPS) is a satellite-based navigation system that is widely used for different navigation applications. In an open sky, GPS can provide an accurate navigation solution. However, in urban canyons, tunnels and indoor environments that block the satellite signals, GPS fails to provide continuous and reliable coverage.
In the search for a more accurate and low-cost positioning solution in GPS-denied areas, researchers are developing integrated navigation algorithms that utilize measurements from low-cost sensors such as accelerometers, gyroscopes, speedometers, barometers and others, and fuses them with the measurements from the GPS receiver. The fusion is accomplished using either Kalman filter (KF), particle filter or artificial-intelligence techniques.
Once a navigation algorithm is developed, verified and proven to be worthy, the ultimate goal is to put it on a low-cost, real-time embedded system. Such a system must acquire and synchronize the measurements from the different sensors, then apply the navigation algorithm, which will integrate these aligned measurements, yielding a real-time solution at a defined rate.
The transition from algorithm research to realization is a crucial step in assessing the practicality and effectiveness of a navigation algorithm and, consequently, the developed embedded system, either for a proof of concept or to be accepted as a consumer product. In the transition process, there is no unique methodology that designers can follow to create an embedded system. Depending on the platform of choice—such as microcontrollers, digital signal processors and field-programmable gate arrays—system designers use different methodologies to develop the final product.
|
|
|
|
|
Navigate to related information
Dr DSP
3/29/2011 2:42 PM EDT
This was an informative article. Maybe EETimes can find a way to host more master project descriptions. Anyone else think that's a good idea?
Sign in to Reply
Carlos1966
3/30/2011 8:05 AM EDT
I wonder what's responsible for the pronounced bias in the error?
Sign in to Reply
Robotics Developer
3/30/2011 6:51 PM EDT
I am wondering if a DSP specific device would provide the same or better bang for the buck? Most of the sensor processing is the filter and that should be implemented very well with a DSP type of controller chip (possibly lower cost as well). Having worked on navigation systems using only accelerometers, gyros, wheel encoders I can say there are many sources of error. The greatest is external forces that exceed the sensors operating range (read here: 3g bump with a 2g sensor). Frequency response, sampling rates(and sampling rate errors), sample errors all contribute to the overall system error. Yes the Kalman filter will help significantly but there are limits.
Sign in to Reply