| Both sides previous revisionPrevious revisionNext revision | Previous revision |
| first:adis16470_imu_frc [09 Dec 2018 22:14] – add driver link Kristen Chong | first:adis16470_imu_frc [05 Jan 2020 21:51] (current) – fix description language which previously was copied from the 448 page Kristen Chong |
|---|
| | <note warning>This page is still under construction for the 2020 season! Please check back here for more updates.</note> |
| ====== ADIS16470 IMU Board for FIRST Robotics ====== | ====== ADIS16470 IMU Board for FIRST Robotics ====== |
| The [[adi>ADIS16470]] IMU board is ADI's next-generation IMU designed for use in a wide variety of applications, including industrial robots, smart agriculture, and autonomous vehicles. The [[adi>ADIS16470]] is a 6-degree-of-freedom IMU with 3-axis gyro and 3-axis accelerometer outputs. This IMU boasts better performance across the board than the [[adi>ADIS16448]], making it a replacement for the IMU board offered in previous years. If your team has used the [[adi>ADIS16448]] IMU board in the past, you'll find this board to operate in a similar manner. | The [[adi>ADIS16470]] IMU board is ADI's rugged, tactical IMU designed for use in a wide variety of applications, including industrial robots, smart agriculture, and autonomous vehicles. The [[adi>ADIS16470]] is a 6 degree-of-freedom IMU, with a 3-axis gyro and a 3-axis accelerometer. |
| \\ | \\ |
| \\ | \\ |
| Supporting code for the [[adi>ADIS16470]] can be found [[https://github.com/juchong/ADIS16470-RoboRIO-Driver|here]], and only exists in the three main programming languages supported by FIRST. If your team wants to use this board with other programming languages, please refer to the [[adi>ADIS16470]] product datasheet for more information on how to communicate with this IMU via SPI. | [[adi>ADIS16470]] software and examples for Java, C+ +, and LabVIEW can be found [[https://github.com/juchong/ADIS16470-RoboRIO-Driver|here]]. If your team wants to use this board with other programming languages, please refer to the [[adi>ADIS16470]] product datasheet for more information on how to communicate with this IMU via SPI. |
| \\ | \\ |
| \\ | \\ |
| \\ | \\ |
| \\ | \\ |
| | {{ :first:adis16470_rotation_figure.jpg?nolink&250|}} |
| ===== Getting Started ===== | ===== Getting Started ===== |
| The ADIS16470 IMU Board is designed to plug directly into the SPI port, just like our simple FRC Gyro Board. However, this board has a "tail" to screw it down to the RoboRIO to prevent it from rattling loose in the middle of a match. You can safely put this "tail" underneath any other boards you may be using on the MXP connector. | The [[adi>ADIS16470]] IMU Board is designed to plug directly into the SPI port on the RoboRIO. The IMU board has an extension with two mounting holes which should be used to secure it to the RoboRIO with #4-40 screws. The figure to the right illustrates the locations of the X, Y, and Z axes in relation to the device package. |
| \\ | \\ |
| | ===== A Note on Offset Calibration ===== |
| | To help the robot minimize start-up drift and improve overall sensor performance, an offset calibration function has been built into the IMU driver. This function captures several seconds worth of data and calculates an average offset which is then applied to the sensor outputs. By default, calibration is automatically started once the RoboRIO begins executing user code and usually takes about 10 seconds to complete. The "Ready?" LED indicator will also illuminate once calibration is complete. New in 2020, all three language libraries also have this functionality made available via the "Recalibrate" function. |
| \\ | \\ |
| ===== A Note on Gyro Calibration ===== | <note important>The gyros used in the [[adi>ADIS16470]] measure //angular rate//, not //angle!// Any movement during the offset calibration routine **will** introduce some error into **every** sensor measurement! Over time, this error, will appear as "drift" in your angle measurement. **It is VERY important that the robot remains completely stationary during this calibration period!**</note> |
| To help the robot identify a starting angle position more accurately and remove gyro drift, it is recommended that you allow the robot to perform what is called an offset calibration first. It will do this by taking several readings from the sensor and averaging them together to obtain a value which will be subtracted from every subsequent reading after the calibration is complete. Calibration typically takes about 30 seconds. The ADIS16470 IMU Board also features an LED which will illuminate when calibration is complete (if used with the Analog Devices supplied libraries). | |
| \\ | \\ |
| | <note tip>Offset calibration should be performed as soon as the robot is powered on to prevent the routine from interfering with any autonomous code execution. If your gyro angle readings are drifting drastically, clicking **Restart Robot Code** in the driver station will force the RoboRIO to re-execute the offset calibration routine. This should fix any drift issues caused by a bad offset measurement. You can also use the new Recalibrate function available in the 2020 libraries to re-perform the offset calibration (if, for example, the robot was kicked or moved during start-up). This function will also perform a "reset," or re-zero the sensor, so be sure the robot is facing in the direction you want "zero" to be!</note> |
| \\ | \\ |
| <note important>Remember, the gyros used in the ADIS16470 are rate sensors, not raw angle sensors, so any movement during this calibration routine will cause the robot to add the measurements of that movement to the value that is subtracted from all readings. ** This will cause a large amount of apparent "drift" in your subsequent sensor readings. Because of this, it is imperative that the robot is left completely stationary during this calibration period!**</note> | |
| \\ | \\ |
| | ===== "Reset" vs. "Recalibrate" ===== |
| | As mentioned above, the 2020 versions of the libraries include a new function called "recalibrate" that takes advantage of features available on the ADIS16470. It's important to note the difference between "reset" and "recalibrate" as they are very different functions and have differing use cases! |
| \\ | \\ |
| Calibration should be performed as soon as the robot is powered on, to prevent the calibration routine from interfering with any autonomous code execution. You can usually tell that your robot was bumped during calibration if you notice that your gyro angle readings are drifting off into space very rapidly. In LabVIEW, this calibration is done as part of the initialization. But in other languages, you must explicitly call the calibrate routine. You can read more info on your team's specific language at the links below. | |
| \\ | \\ |
| | ==== Recalibrate ==== |
| | "Recalibrating" the IMU will reconfigure the offset calibration in the device. Upon startup, the libraries for the ADIS16470 will automatically perform an offset calibration as described above. It does this by taking readings for a few seconds (the default is 4 seconds) and averaging those values together to form an offset, one for each axis. This allows the IMU to calibrate out **some** drift to improve the sensor performance. We discussed above that any movement of the robot during this calibration period will negatively impact sensor performance. Before 2020, there was no function that would allow teams to recalibrate the IMU without restarting robot code. If the robot was bumped or moved during startup, IMU performance would be impaired. The ADIS16470 has an internal feature which continuously averages the internal sensor readings and stores them on the device. The device can be instructed to use these averages as the new internal offset values when a certain command is issued. This is what the new recalibrate function does. |
| \\ | \\ |
| ===== Using the Gyro on Your Robot ===== | <note important>The Recalibrate function should **never** be used during match play! Doing so may negatively affect sensor performance if the robot moves before Recalibrate is called, just as bumping the robot during the startup calibration will negatively affect sensor performance!</note> |
| For more information on how to add gyro functionality to your robot code, select your team's preferred programming language from the list below. | \\ |
| | ==== Reset ==== |
| | "Resetting" the IMU simply "re-zeros" the device. You may recall that gyros output data as degrees per second, not degrees,and we must integrate the readings from the device in order to determine a heading in degrees. When we reset the IMU, we are resetting this integration and starting back at zero. This means that the direction that the robot is facing when you reset the IMU will become the new zero degrees direction. |
| | \\ |
| | \\ |
| | ===== Using the ADIS16470 IMU on Your Robot ===== |
| | For more information on how to add IMU functionality to your robot code, select your team's programming language from the list below. |
| \\ | \\ |
| \\ | \\ |
| [[first:adis16470_imu_frc:labview|Using the ADIS16470 IMU in LabVIEW]] | [[first:adis16470_imu_frc:labview|Using the ADIS16470 IMU in LabVIEW]] |
| \\ | \\ |
| [[first:adis16470_imu_frc:cpp|Using the ADIS16470 IMU C++]] | [[first:adis16470_imu_frc:cpp|Using the ADIS16470 IMU in C++]] |
| \\ | \\ |
| [[first:adis16470_imu_frc:java|Using the ADIS16470 IMU in Java]] | [[first:adis16470_imu_frc:java|Using the ADIS16470IMU in Java]] |
| |
