MAVS Lidar Model

The lidar sensor can also be accessed through the python interface.

Creating a MAVS Lidar

A lidar can be created in one of three ways. First, one of the built-in lidar models can be used.

mavs::sensor::lidar::Lms291_S05 s05;
mavs::sensor::lidar::Vlp16 vlp;
mavs::sensor::lidar::Hdl32E hdl32;
mavs::sensor::lidar::Hdl64E hdl64;
mavs::sensor::lidar::MEight m8;
mavs::sensor::lidar::OusterOS1 os1;
mavs::sensor::lidar::OusterOS2 os2;
mavs::sensor::lidar::Rs32 rs32;

Secondly, a lidar can be created by explicitly setting the scan properties. See the lidar API documentation for more information.

mavs::sensor::lidar::Lidar lidar;
lidar.SetScanPattern(-180.0f,179.0,1.0f,-15.8f,15.8f,1.0f);

A lidar can also be created by loading a lidar input file.

mavs::sensor::lidar::Lidar lidar;
lidar.Load("path/to/mavs/data/sensors/LMS291.json");

An example input file is

{
  "Scan Pattern": {
    "Horizontal Range": [-180,179.75],
    "Horizontal Step": 0.25,
    "Vertical Range": [-15.8,15.8],
    "Vertical Step": 1.0
  },

  "Mode": 1,

  "Beam Properties":{
    "Shape": "circle",
    "Divergence": [0.012,0.012]
  },

  "Max Range": 80.0,
  "Min Range": 0.0
}

The parameters are defined below. Not that these parameters can also be set through the API.

• Max Range: The maximum range of the lidar in meters
• Min Range: The minimum range of the lidar in meters
• Beam Properties:
  • Shape: Can be “circle”, “rectangle”, or “ellipse”
  • Divergence: The beam divergence in the horizontal and vertical planes, in radians/m
• Scan Pattern:
  • Horizontal Range:
  • Horizontal Step:
  • Vertical Range:
  • Vertical Step:
• Mode: The return mode of the lidar, as an integer value from 0-3. The default mode is 1
  • 0: Average return distance
  • 1: strongest return
  • 2: last return
  • 3: strongest and last return

Example Lidar Simulation

The following example ilustrates many of the features of the MAVS lidar simulation. More explanation of the function calls can be found in the lidar API documentation.

// Include a ray tracer and lidar model
#include "raytracers/simple_tracer/simple_tracer.h"
#include "sensors/lidar/vlp16.h"

int main (int argc, char *argv[]){
  // Create a temple test scene
  mavs::raytracer::SimpleTracer scene;
  scene.CreateTestScene();
  // Create an environment and add the scene to it
  mavs::environment::Environment env;
  env.SetRaytracer(&scene);
  // Specify the pose / state of the sensor
  glm::vec3 sens_pos(-25.0f, 0.0f, 10.0f);
  glm::vec3 sens_vel(1.0f, 0.0f, 0.0f;
  glm::quat sens_orientation(1.0f, 0.0f, 0.0f, 0.0f);
  // Create the lidar sensor to simulate
  mavs::sensor::lidar::Vlp16 vlp;
  // Specify the time step. 
  //Should be the revolution time of the sensor
  float dt = 0.1f;
  // Set the position & orientation of the sensor
  lidar->SetPose(sens_pos, sens_orientation);
  // Set the velocity of the sensor
  // This will automatically calculate movement during the scan
  lidar->SetVelocity(sens_vel.x, sens_vel.y, sens_vel.z);
  // Update the sensor and point cloud
  lidar->Update(&env, dt);

  // display a top-down view of the point cloud
  // points will be rgb colorized
  // options are "color", "range", "height", "intensity", and "white"
  lidar->SetDisplayColorType("color");
  lidar->Display();
  // Now display perspective view by height
  lidar->SetDisplayColorType("height");
  lidar->DisplayPerspective();
  // Save the point cloud to a labeled PCD file
  lidar->AnnotateFrame(&env,true);
  lidar->WritePcdWithLabels("labeled_points.pcd);
  return 0;
}

Updating the MAVS Lidar