The attached is the windows installer for the latest verson of the Windows version of the Freescale Sensor Fusion Toolbox.  This version must be used with Version 5.00 of the sensor fusion library.  It is NOT backward compatible, as the structure of the undocumented Kalman filter packet has changed. Please uninstall any prior versions of the toolbox before running this installer. As in prior releases, you can reflash your Freedom boards from within the application itself.  So if you want to try the new fusion without downloading the new library and firing up KDS or CodeWarrior, you can. Version 5.00 was posted to this same board just a few minutes ago in a separate post.  It contains full documentation, including algorithm development, datasheet and user guide. Regards, Mike Stanley

Attached is a PDF dump of a classic blog post.  Please forgive the format problems.  I'll work with the web team in January to get it reposted to the NXP site. Mike

As requested in a prior posting...

I am occasionally asked for the source code for the Android version of our Sensor Fusion Toolbox.  Well, here it is, complete with new NXP graphics and updated links.

All, We've had some problems with migration of Freescale blogs over to the NXP domain.  One set of listings that gets requested a fair bit are the postings on orientation representations.  The attached is a PDF version of the same material which will be contained in the upcoming Sensor Fusion Version 7.00 User Guide.

You saw it here first! The attached zip file contains full documentation and source code (both CodeWarrior and KDS) for Version 5 of Freescale's Sensor Fusion Library for Kinetis MCUs. The 6 and 9-axis Kalman filters have been rewritten from scratch by fusion expert Mark Pedley.  The new versions have improved dynamic tracking and at-rest stability.  All documentation has been updated, and full implementation details are included.  We continue to use the BSD 3-clause software license, so you are virtually unlimited in how you use the library.   I will note that due to the Kalman filter changes, Mark made some changes to the undocumented Kalman filter packets which are utilized by the Windows Version of the Freescale Sensor Fusion Toolbox.  This breaks compatibility with older versions of the toolbox.  I will be posting a pre-release of the new one immediately after completing this post.  The Android version of the toolbox does not make use of the Kalman filter packet, and should be forward compatible - although an updated version of that is in the works as well.   Structure of the new library is very similar to Version 4.22, which was the previous production version.  There are changes, but I don't expect anyone to have problems adapting existing projects to the new version of the library.  I think you will get improved performance if you do.   Regards, Mike Stanley

This is a pre-release version of the Sensor Fusion Toolbox for Windows.  It is essentially the same tool you already know and love.  It differs in terms of the binary images that can be downloaded to your development board.  Clicking File->Flash from the toolbar will present you with a set up updated binaries.  These were built using a brand new set of 6- and 9-axis Kalman filter algorithms.  We expect these to consume much less power (2X or 3X) and to have better gyro tracking.  We are soliciting feedback from existing users of the toolkit with regard to performance of the new filters.  We do not expect to formally release until later in Q2, primarily to give the community time to try them and give us feedback.  We are not releasing source just yet, for the same reason.

For those of you who do not have access to Google Play, here is the latest .apk file for the Android version of the Sensor Fusion Toolbox.  This version should trap problems which caused previously reported crashes.  There are no visible changes to the functionality.   IF you have access to Google Play, we recommend that you use that as your default installer, as you can automatically get updates.

Posted here in response to a query by Michael Smorto. Regards, Mike Stanley

This presentation was given at the October 2014 ARM TechCon in Santa Clara, CA.  It provides a detailed description of recent changes to Freescale's sensor fusion offering.

This this shows how to implement the power cycling feature described in Section 4.8, Fusion Standby mode, of the Version 7.00 Sensor Fusion User Guide.  It will power down the gyro when the board is stationary, and also suspend sensor fusion.   Last computed results continue to be sent until new motion is detected.  One nice side effect is that 6-axis yaw drift is almost eliminated.

The attached is a preview copy of build 422 of the sensor fusion library, which is currently being tested by the Freescale Sensor's team. Please consult the docs/Release_Notes.txt file for changes from build 420, as well as known errata. This version is released under the following license:   Freescale Sensor Fusion Library for Kinetis MCUs IMPORTANT. Read the following Freescale Software License Agreement (“Agreement”) completely.  By downloading this file, you indicate that you accept the terms of this Agreement and you also acknowledge that you have the authority, on behalf of your company, to bind your company to such terms.  You may then download or install the file. FREESCALE END-USER SOFTWARE LICENSE AGREEMENT Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:     * Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.     * Redistributions in binary form must reproduce the above copyright  notice, this list of conditions and the following disclaimer in the  documentation and/or other materials provided with the distribution.     * Neither the name of Freescale Semiconductor, Inc. nor the  names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL FREESCALE SEMICONDUCTOR, INC. BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. © 2004-2014 Freescale Semiconductor, Inc. All rights reserved.

All, This is my personal "cheat sheet" that I use as reference whenever I have to code angular transformations from one frame of reference to another.  There's nothing unique here, it just organizes things in a way that I can find them quickly.  I hope you find it useful. Mike

Here's a zip file which incorporates the patch I outlined in my previous posting.

All, The attached was put together in response to the posting by Andrew Hartnett.  It contains a bare-metal IAR project for 9-axis sensor fusion V7.00 on the KL25Z.  You need to have built KSDK for the KL25Z to include the ISSDK option.  Then unzip this file into your SDK_2.0_FRDM-KL25Z/boards directory.  The sample project is then located at SDK_2.0_FRDM-KL25Z/boards/frdmkl25z_virtual_shield/issdk_examples/algorithms/sensorfusion/baremetal_sensor_fusion/iar. There is also an included freertos_sensor_fusion project.  Ignore that for now.  It compiles and links, but needs more RAM than the KL25Z supplies.  I'm looking at ways to decrease the RAM requirements to fit. Regards, Mike

Attached is an LPCExpresso project for LPC1549.  It is compatible with the latest version of the Sensor Fusion Toolbox for Windows (the version targeted at Version 6.00 and 7.00 sensor fusion).  This project is a variant on the Sensor Fusion Version 6.00 library.  Algorithmically this is virtually identical to Version 7.00.

Android sensor fusion APK

You will have to add a .exe extension to the unzipped file.