Advait Jain (http://www.cc.gatech.edu/~advait)
What this does
Tilting the Hokuyo using a servo and reading back the servo encoder position while getting scans from the Hokuyo allows us to generate a 3D point cloud of the world in front of the robot. We have used this 3D point cloud for object segmentation, door handle detection etc.
We first used this sensor in this paper: EL-E: An Assistive Mobile Manipulator that Autonomously Fetches Objects from Flat Surfaces. Advait Jain and Charles C. Kemp. Autonomous Robots, 2009. This paper and others can be downloaded from the Heathcare Robotics Lab website (http://www.healthcare-robotics.com)
Pictures of the Setup
This is the hardware setup that we use:
- Tilting servo: Robotis Servo (typically we use RX-28) http://www.crustcrawler.com/motors/RX28/index.php?prod=66
- This gives a serial interface to control the servo: http://www.crustcrawler.com/electronics/USB2Dynamixel/index.php?prod=65
- Hokuyo UTM. (http://www.acroname.com/robotics/parts/R314-HOKUYO-LASER4.html)
- Bracket to mount the Hokuyo on to the servo.
- This is a custom part that we have designed.
- The solidworks CAD file (utm-servo-bracket.sldprt). This requires Solidworks 2009.
- We have taken two routes for making the bracket
- We thank Abhishek Bhatkhande for making the CAD model from the bracket design.
- ROS (http://www.ros.org)
- You will need the following ROS packages from our public ROS repository (available at http://code.google.com/p/gt-ros-pkg/wiki/hrl_content_summary)
- hrl_hokuyo (requires hokuyo_node - http://www.ros.org/wiki/hokuyo_node)