BerryGPS-IMU GPS and 10DOF for The Raspberry Pi - accelerometer, gyroscope, magnetometer and barometric/Altitude sensor

I ran the IMU at 100Hz, though adjustable (I recall ... been a while). I ran the GPS at 1Hz. Only problem from my perspective is that the IMU and GPS are disconnected, that is, the supplied IMU code does not query the GPS. I found it difficult to coordinate the timing of the IMU and GPS. Time stamps are not common.

The accelerometer, gyroscope, and GPS work great. Unfortunately, even after extensive troubleshooting, I could not get reliable readings from the magnetometer. After reading through the reviews, there is a high probability of receiving a defective unit.

There are numerous IMU and GPS solutions for the raspberry pi. What sets this product apart are the following ... 1. IMU and GPS are integrated into a single module. Many other solutions require separate IMU and GPS chips/boards, which puts more burden on the user to figure out connections to the 40 pin header. The BerryGPS-IMUv3 integrates everything and consolidates the interfaces to the first 10 pins of the header. It also has through-holes nearby that make these pins available for other purposes if a user needs these for another device. In short, a thoughtful design. 2. The module also includes a common Bosch pressure transducer that provides measurements to infer altitude for an altimeter. Together with the 9-axis (accel/gyro/mag) this gives a nice 10-DOF capability. Again, one can add the pressure transducer separately to get the same capability, but the integration into the single board makes this much more plug-and-play from a mechanical/pin-use standpoint. 3. The module is compact! Given all of the sensors/chips included, it takes up much less space than if one assembled a system from the other discrete IMU, GPS, and barometric pressure sensors. 4. The manufacturer maintains a website and associated forums. The documentation here really is what sold me in addition to the other solutions at similar price points; those alternatives basically consist of pointing you to datasheets. To be fair, AdaFruit modules have a nice circuitpython library to have software support for its sensors. However, many of those can be used for this module too, as it uses some similar chips. (I preferred to ultimately write my own for efficiency/correctness of some of the nuanced details, but the main point is that documentation and software resources is another strong point of the BerryGPS-IMUv3 compared to other offerings.) 4. The GPS module supports either the internal antenna on the u-blox M8 GPS receiver or an external active antenna. NB: It was not apparent to me when ordering that the external antenna is required to be an active antenna. To use a passive antenna you need to make a small mod to board. I had a question about this and the manufacturer (Ozzmaker) was very responsive and helpful. 5. Great support. See #4 above. All in all, a great module for $50 US.

Tried to connect this to my Arduino using I2C. Could not get a connection. I wish they never advertised this for arduino. I also could not find a library that would work with the module its supposed to be berryIMU.h or something like that. Definitely not for arduino stay away!

Everything is great. Only bonus would be if PPS pin was wired up out of the box.

I ordered one of these only to find that the GPS function worked, but not anything on the I2C bus. I ordered a second unit on a whim that there was a hardware problem. Second unit has worked perfectly so far. I have outfitted in my waverunner with a raspberry pi. The pi records the IMU data, pressure, temperature, GPS, and AIS and stores it onboard. I hope unit number two has some longevity.

I picked up this device to start to learn about these little I2C IMU sensors and get a feel for what they're capable of. The attached image shows the Grafana dashboard I use to visualize the data. I wrote a script that collects data and feeds it to logstash then on to Kafka and Elasticsearch to quench my data science thirst. Next steps I think are to read up on smoothing the data, because the raw data seems to have a good amount of noise. From what I can tell it's electromagnetic or electrostatic noise. I have mine mounted very firmly to the cinder block basement wall so it shouldn't be showing so much movement. With that said, the noise level is somewhere around 1/1000th of a G, so this level of precision is extremely well suited for just about any sort of robotic/drone/aircraft application you can think of.

I wasn't able to get a reading from the BMP280 chip.. temp/pressure. It shows up on the I2Cbus (0x77), but when you dump all the bytes, even the ID comes back as 0x00. I may try to cut the trace on the back to put it on 0x76, but i have doubts that will change anything. Other than that, everything else worked. GPS required quite a clear LOS to the sky with the internal antenna.