For a visualization of a simplified model of our excavator HEAP clone the following repository into the workspace:
https://github.com/leggedrobotics/rsl_heap.git
This step can also be left out.
This is the example of the code of our ICRA2022 paper [1]. For compiling this code simply run:
catkin build excavator_dual_graph
For two example scenarios on our test-field at ETH Zurich please download the data located at the Google Drive Link.
In terminal 1 run (after sourcing the workspace):
roslaunch excavator_dual_graph dual_graph.launch
# or if heap_urdf not available
roslaunch excavator_dual_graph dual_graph_no_urdf.launch
In terminal 2 play one of the rosbags located at the provided link location. Make sure to set the --clock
-flag.
Either
rosbag play 1_digging_filtered.bag --clock
# or
rosbag play 2_driving_away_filtered.bag --clock
The expected output looks like this:
Here, yellow depicts the measured GNSS path, green depicts the estimated path in the world frame, and blue depicts the estimated path in the odometry frame.