Lightware SF45/B Rotary Lidar

LightWare SF45/B is an extremely small and light rotating Lidar with a range of 50m. It can be used to enable Collision Prevention without a companion computer.

INFO

The lidar driver is not included in the default build of PX4. You will need to create and use a custom build.

LightWare Studio Setup

In the LightWare Studio app set following values:

Параметр	Опис
Baud rate	921600

Make sure the scan angles are set so that nothing on the drone interferes with the measurements. The driver and Collision Prevention automatically handle angles different from the maximum angles.

Налаштування програмного забезпечення

The rangefinder can be connected to any unused serial port, such as TELEM2. Parameter Configuration explains how to configure the port to use and the other properties of the rangefinder.

Налаштування PX4

Add the Driver to the PX4 Build

The driver for this LiDar is not included in PX4 firmware by default.

Вам потрібно:

1. Add the lightware_sf45_serial driver to firmware:
   • Install and open menuconfig
   • In menuconfig, navigate to Drivers > Distance sensors
   • Select/Enable lightware_sf45_serial
2. Build PX4 for your flight controller target and then upload the new firmware.

Налаштування параметрів

You will need to configure PX4 to indicate the serial port to which the sensor is connected (as per Serial Port Configuration) and also the orientation and other properties of the sensor.

The parameters to change are listed in the table.

Параметр	Опис
SENS_EN_SF45_CFG	Set to the serial port you have the sensor connected to.
SF45_ORIENT_CFG	Set the orientation of the sensor (facing up or down)
SF45_UPDATE_CFG	Set the update rate
SF45_YAW_CFG	Set the yaw orientation

Тестування

You can confirm that the sensor is correctly configured by connecting QGroundControl, and observing that OBSTACLE_DISTANCE is present in the MAVLink Inspector.

Налаштування перешкод в QGC буде виглядати так:

Driver Implementation

The sensor driver publishes the ObstacleDistance UORB Message that is used by PX4 Collision Prevention. The measurements in each sector will correspond to the lowest measurement the sensor had in that corresponding sector. The data is then published to the OBSTACLE_DISTANCE MAVLink message.

Усунення проблем

Errors and Jerky Movements

Start-up issues, jerky movements, or a lot of communication errors displayed with lightware_sf45_serial status could mean that the sensor is not getting enough power.

According to its datasheet, the sensor needs 300 mA of current at 5V. The supplied cable is fairly long and when connected to the 5V supply, such as the flight controller TELEM2 port, the voltage at the rangefinder may drop below the required level. One mitigation/alternative is to power the SF45 via a separate step-down converter from battery voltage, ensuring that there are 5V across the rangefinder itself.

Debugging with PlotJuggler

Debugging problems with distance sensors is much easier if you can plot the OBSTACLE_DISTANCE message in an x-y plot, because this lets you directly see the orientation of the measurement in respect of the drone.

The video below shows how you can view such a plot in PlotJuggler.

In order to generate this kind of plot you will need to add the following reactive Lua scripts to PlotJuggler. If you save these scripts and then add new data, you will see a new timeseries called obstacle_distance_xy, which is the same as is displayed in the video.

TIP

The Plotting Obstacle Distance and Minimum Distance in Real-Time with PlotJuggler in the Collision Prevention topic has a more detailed example of how use Reactive scripts in PlotJuggler.

Global code, executed once:

obs_dist_xy = ScatterXY.new("obstacle_distance_xy")

function(tracker_time)

obs_dist_xy:clear()

i = 0
angle_offset = TimeseriesView.find("obstacle_distance/angle_offset")
increment = TimeseriesView.find("obstacle_distance/increment")

while(true) do

str = string.format("obstacle_distance/distances.%02d", i)
distance = TimeseriesView.find( str )

if distance == nil then break end
angle = angle_offset:atTime(tracker_time) + i * increment:atTime(tracker_time)

y = distance:atTime(tracker_time) * math.cos(math.rad(angle))
x = distance:atTime(tracker_time) * math.sin(math.rad(angle))

obs_dist_xy:push_back(x, y)

i = i + 1
end

TIP

To also see the fused obstacle_distance, you can just adapt the script to work on the obstacle_distance_fused message.