Using SaDVIO Library with your own Dataset

This tutorial explains how to use the SaDVIO library with your own dataset. It covers the necessary steps to prepare your data, create a proper calibration file, and run the library. Introduction

To use SaDVIO with your own dataset, you need to satsify the following:

• A dataset containing synchronized image measurements (+ IMU)
• A calibration file describing the intrinsic and extrinsic parameters of your sensor suite.

Dataset Format

The configuration of your dataset should be stored in .yaml file, itself stored in a folder containing a dataset folder and a config.yaml file. The config.yaml file contains all the SLAM related parameters, and the dataset folder should contain a bunch of .yaml file that describes each dataset. For instance, the eth.yaml file (e.g. for the EUROC dataset) looks like:

# General sensor definitions.
dataset: EUROC
ncam: 2
 
camera_0:
  topic: /cam0/image_raw
  # Sensor extrinsics wrt. the body-frame.
  T_BS:
    cols: 4
    rows: 4
    data: [0.0148655429818, -0.999880929698, 0.00414029679422, -0.0216401454975,
          0.999557249008, 0.0149672133247, 0.025715529948, -0.064676986768,
        -0.0257744366974, 0.00375618835797, 0.999660727178, 0.00981073058949,
          0.0, 0.0, 0.0, 1.0]
 
  # Camera specific definitions.
  rate_hz: 20
  resolution: [752, 480]
  projection_model: pinhole
  intrinsics: [458.654, 457.296, 367.215, 248.375] #fu, fv, cu, cv
  distortion_model: radial-tangential
  distortion_coefficients: [-0.28340811, 0.07395907, 0.00019359, 1.76187114e-05]
 
camera_1:
  topic: /cam1/image_raw
  # Sensor extrinsics wrt. the body-frame.
  T_BS:
    cols: 4
    rows: 4
    data: [0.0125552670891, -0.999755099723, 0.0182237714554, -0.0198435579556,
          0.999598781151, 0.0130119051815, 0.0251588363115, 0.0453689425024,
      -0.0253898008918, 0.0179005838253, 0.999517347078, 0.00786212447038,
          0.0, 0.0, 0.0, 1.0]
 
  # Camera specific definitions.
  rate_hz: 20
  resolution: [752, 480]
  projection_model: pinhole
  intrinsics: [457.587, 456.134, 379.999, 255.238] #fu, fv, cu, cv
  distortion_model: radial-tangential
  distortion_coefficients: [-0.28368365,  0.07451284, -0.00010473, -3.55590700e-05]
 
imu:
  topic: /imu0
  T_BS:
    cols: 4
    rows: 4
    data: [1.0, 0.0, 0.0, 0.0,
          0.0, 1.0, 0.0, 0.0,
          0.0, 0.0, 1.0, 0.0,
          0.0, 0.0, 0.0, 1.0]
  rate_hz: 200
 
  # inertial sensor noise model parameters (static)
  gyroscope_noise_density: 1.6968e-04     # [ rad / s / sqrt(Hz) ]   ( gyro "white noise" )
  gyroscope_random_walk: 1.9393e-05       # [ rad / s^2 / sqrt(Hz) ] ( gyro bias diffusion )
  accelerometer_noise_density: 2.0000e-2  # [ m / s^2 / sqrt(Hz) ]   ( accel "white noise" )
  accelerometer_random_walk: 3.0000e-1    # [ m / s^3 / sqrt(Hz) ]  ( accel bias diffusion )
  dt_imu_cam: 0.0

So here you need to know precisely the transform between every sensor, as well as their intrinsic properties. A wonderfull toolbox that you can use for such a task is kalibr

Run Without ROS2

If you don't use the middleware ROS2, you can use the EUROC dataset ASL data format for which a data grabber is already implemented in ADataprovider.h. Then you can compile and run the executable main.cpp in the /cpp as such

mkdir build
cd build
cmake ..
make

And to run the executable

./isaeslam "/your/path/SaDVIO/ros/config" "/your/path/to/your/dataset/in/ASL/format"

Run with ROS2

With ROS2