Rotating XRP Rail (cheap lidar alternative)

Rotating XRP Rail

Take anything that mounts on the rail of the XRP and rotate it using a servo. Don’t be fooled by its simplicity. This 50mm rotating XRP rail enables some cool stuff in robotics.

Get it

• (1) 9g servo (it should come with a servo horn and screw)
• No tools required, but a screwdriver is recommended

Print it

• PLA
• 0.2mm layer height
• 15% infill
• No supports

Build it

Step 1: Attach the servo horn to your servo.
Step 1b [optional]: Use a screwdriver to secure the horn to the servo.
Step 2: Insert the servo horn into the rail. It should snap into place.
Step 3: Insert the servo into the servo holder
Step 4: Attach the whole mechanism to your XRP

Code it

Step 1:

Download xrp_rotate_rangefinder_example.py from this Github repository and run it on your XRP using https://xrpcode.wpi.edu/ . Push the “USER” button to sweep the ultrasonic sensor.

Step 2:

Copy your data array from the terminal and paste it into plot_rangefinder_data.py from this Github repository. Run it on a machine with matplotlib installed, or run it online here (paste the code on the left side, and then click the Preview button on the top left).

Here’s a real example of what a room looks like to an ultrasonic sensor:

Teach it

This rotating rail gives you more sensor data per robot pose so you can teach more about robotics with just dead reckoning.

Lesson ideas:

• Effects of material properties on different kinds of sensors
• Coordinate frames and transformations
• Occupancy grids and uncertainty
• Probability
• Visualizing data with matplotlib

Project ideas:

• Mount a ToF distance sensor (use this mount) and now you have a budget “planar lidar”.
  • Compare range readings from different materials (matte black, reflective, transparent)
  • Sweep the sensor past a sharp corner with another object behind it to observe the veiling effect. Learn to filter the data like the ScanShadowsFilter in the laser_filters package in ROS
• Mount an ultrasonic sensor (use the “Peripheral - Ultrasonic - Clip.3MF”)
  • Compare range readings from different materials and with different orientations (soft materials, rigid surfaces angled out of plane, multipath)
  • Drive the robot in a square, take measurements at each point, and use a conical sensor model plus probability to fill out an occupancy grid.

History

• V1: initial release