MIMU-Toolbox

The RAW output format

Each row group of 6: 1, x axis acc. 2, y axis acc. 3, z axis acc. 4, x axis gyro 5, y axis gyro 6, z axis gyro

Transfer rate

Bluetooth

Hard to get streaming up to 1000 Hz when using normal IMU

USB

Can get 1000 Hz, but the period time will jitter.

Collection of MATLAB functions for reading out IMU data.

Devices

The new one is MIMU22B96v1, the old is MIMU22BTv2

Single READOUT is unstable

Does not know if there is a measurement, could all be zero

Then the MIMU continues to output packages, even though stop command is sent

MPU (Micro Processing Unit)

Should be AT32UC3C2512C. https://www.microchip.com/wwwproducts/en/AT32UC3C2512C#datasheet-toggle

Algorithms

IMU_TS_SID // Should be the timestamp?

Variables

u->f.x is the specific force, accelerometer reading in x
u->g.x is the angular velocity.

The IMU readings is stored in the variable extern int16_t mimu_data[32][10]; where 32 is the number of IMUs. Second position is

1. imu[0] -> acc.x
1. imu[1] -> acc.y
1. imu[2] -> acc.z
1. imu[3] -> gyro.x
1. imu[4] -> gyro.y
1. imu[5] -> gyro.z

Endianess

The endianess on linux computer is using lscpu | grep "Byte Order"

Byte Order:            Little Endian

But the data comes with Big endian

Baudrate

To view the current Baudrate

stty -F /dev/ttyACM0

To change to the maximum allowed in linux (https://github.com/jbkim/Linux-custom-baud-rate)

stty -F /dev/ttyACM0 460800

Should be fine for 4 IMUs

Isaac claims that Baudrate settings does not matter, it is over USB that has Mbit/s.

Read rate

If the BytesAvailableFcnCount is to low, computer will sink down.

Sensor

The old one seems to have 2000 Hz and the new one have 1000 hz? Divide the rate of the old sensor by two, to have equal rates.

Output data

If even rate output is requested for raw imu data, then IMU time stamp is also included.

To read out

Need to execute use_as_normal_imu.m first, otherwise only zeros will be output.

Transformation between IMUs

In the MIMU22, IMUs 1 and 3 ,and IMUs 2 and 4 have different transformation matrices. To rotate to a righthanded coordinate system IMUs 1 and 3 have

[1 0 0; y_raw = y_platform
0 -1 0;
0 0 -1]

Reflection in yz-plane, for some reason.

For imus 2 and 4

[0 -1 0; y_raw = y_platform
1 0 0;
0 0 1]

The above is rotation around z axis 90 deg.

Which is strange. The acc transformation is the same as the gyro transformation, assuming gravity should be measured positively when sensitivity axis points upwards.

This was found from the measurement sequence in temp_1bin and temp_2.bin.

| Gyro | Acc | Normal IMU | Raw 13 | Raw 24 | | | -z | +z | +z | -z | | +x | | +x | +x | -y | | | -y | +y | +y | -x | | -x | | -x | -x | | | | -z | +z | +z | | | -y | | +y | y | -x | | | -x | -x | -x | +y | | +y | | -y | -y | +x | | | -z | +z | | | | -z | | +z | +z | -z | | | -z | +z | +z | | | +z | | -z | -z | +z | | | | | | |

The two left columns follow the coordinate system on the casing of the new array IMU.

Order of execution

Execute use_as_normal_imu.m from OpenShoe repo. Must be done to activate the device, dont know the reason.
Open the devices with open_device_0.m and open_device_1.m, that put com0 and com in workspace.
Execute

Extract the data from the binary files:

 data1 = get_data_binary_file('temp_1.bin');
 data2 = get_data_binary_file('temp_2.bin');

Parse the data with:

Rate Divider

| Divider | Rate (Hz) | | 1 | 1000 | | 2 | 500 | | 3 | 250 | | 4 | 125 | | 5 | 62.5 | | 6 | 31.25 | | 7 | 15.625 |

hcarlsso/MIMU-Toolbox